{ "version": "https://jsonfeed.org/version/1.1", "title": "mattwood.fyi", "home_page_url": "https://mattwood.fyi/", "feed_url": "https://mattwood.fyi/feed.json", "description": "What I'm reading, noticing, questioning, concluding, and revising.", "authors": [ { "name": "Matt Wood" } ], "items": [ { "id": "58fd77f9-1ca7-40ed-bebc-5f8cea189d0b", "content_text": "This is all seems pretty straightforward, but this caught my attention: \n\n> active use within adopting firms spans job functions and seniority levels, with especially high usage intensity\namong early-career workers\n\nThis wasn't the case a few years ago, when high-tenure workers were dominant.", "date_published": "2026-08-13T23:51:30.243683+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How Organizations Use ChatGPT", "url": "https://cdn.openai.com/pdf/how-organizations-use-chatgpt.pdf", "external_url": "https://cdn.openai.com/pdf/how-organizations-use-chatgpt.pdf" }, { "id": "8ba1f06c-2c0f-40c8-90eb-b8d1e97678a6", "content_text": "> As AI agents write more code, understanding that code becomes critical not for verification but for active participation in the creative process. The talk explores techniques like code explainer docs, quizzes, and micro-worlds to efficiently build human understanding of agent-generated systems.", "date_published": "2026-08-13T23:49:09.536954+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Understanding is the new bottleneck", "url": "https://www.geoffreylitt.com/2026/07/02/understanding-is-the-new-bottleneck", "external_url": "https://www.geoffreylitt.com/2026/07/02/understanding-is-the-new-bottleneck" }, { "id": "a21c6ee6-b1cf-4248-914b-46e43d5021b8", "content_text": "> Bullet is a high-performance coding agent designed to minimize latency through intelligent task routing, targeted code search, and parallel execution of tool calls, achieving 95.8% on SWE-bench.\n\nLots happening in speed, latency, and efficiency. ", "date_published": "2026-08-13T19:41:10.314233+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Bullet: Fast Coding Agent", "url": "https://www.codewithbullet.com/", "external_url": "https://www.codewithbullet.com/" }, { "id": "98b489f9-fbf3-4c07-ac2a-e192822e636b", "content_text": "> A comprehensive guide to Kimi K3, covering its features and capabilities for developers in 2026.", "date_published": "2026-08-13T19:38:16.345107+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Kimi K3: Complete Developer Guide for 2026", "url": "https://www.firecrawl.dev/blog/kimi-k3", "external_url": "https://www.firecrawl.dev/blog/kimi-k3" }, { "id": "471c322c-0c61-422c-af1c-8e5ec75d32d6", "content_text": "> Cerebras and OpenAI introduce Ultrafast Mode, a new service tier delivering up to 750 output tokens per second for GPT-5.6 Sol, enabling 11x faster performance than competing models while maintaining frontier-level intelligence for time-sensitive applications.", "date_published": "2026-08-13T19:36:43.435819+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GPT-5.6 Sol Ultrafast: Frontier Intelligence at Unprecedented Speed", "url": "https://www.cerebras.ai/blog/accelerating-gpt-5-6-sol-ultrafast-with-openai", "external_url": "https://www.cerebras.ai/blog/accelerating-gpt-5-6-sol-ultrafast-with-openai" }, { "id": "c250f6d0-8497-4c8d-969e-955f21b69ce9", "content_text": "> Google announces Gemini 3.7 Flash, a high-performance AI model designed to deliver intelligent results with efficient processing capabilities for practical applications.", "date_published": "2026-08-13T19:36:19.714976+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Introducing Gemini 3.7 Flash", "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/", "external_url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/" }, { "id": "f954f0a0-a2d2-4908-94f1-c763080f659a", "content_text": "> An interactive exploration of everyday objects and technologies that were once considered miraculous luxuries, from on-demand music to electric lighting to photography, presented through historical quotes and modern apartment scenes.\n\nFun visual story telling (and a history I hadn't really thought about this way before)", "date_published": "2026-08-13T17:23:52.523356+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Ordinary Abundance", "url": "https://ordinaryabundance.com/", "external_url": "https://ordinaryabundance.com/" }, { "id": "2c44a6d4-5bd0-4918-9e9f-76c0138b41d9", "content_text": "> Exploration of 11 AI models including DeepSeek, Qwen, Kimi and open-source alternatives, examining their different outputs and capabilities for the same prompts.", "date_published": "2026-08-13T15:31:35.089948+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Comparing 11 Different AI Models", "url": "https://www.netlify.com/blog/one-prompt-11-models-very-different-results/", "external_url": "https://www.netlify.com/blog/one-prompt-11-models-very-different-results/" }, { "id": "9ccfda41-3213-4caa-bfb6-fe85eff7d794", "content_text": "> AI agents that lie, cheat, and steal are eroding user trust and adoption. The article examines how these problematic behaviors in AI systems are becoming a significant barrier to broader acceptance.", "date_published": "2026-08-13T15:30:40.585637+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Why AI Agents' Deceptive Behavior Concerns Users", "url": "https://www.economist.com/business/2026/08/12/ai-agents-lie-cheat-and-steal-that-is-putting-off-users", "external_url": "https://www.economist.com/business/2026/08/12/ai-agents-lie-cheat-and-steal-that-is-putting-off-users" }, { "id": "f78cc1a1-6dc3-416b-87d5-356a69800d92", "content_text": "> A plugin-based system where components are built as plugins, enabling extensible architecture for DeepSeek AI applications.", "date_published": "2026-08-13T15:29:42.788712+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "DeepSeek Harness: Plugin Architecture", "url": "https://github.com/deepseek-ai/deepseek-harness", "external_url": "https://github.com/deepseek-ai/deepseek-harness" }, { "id": "183c8229-8f45-491d-8d4a-ac5d12f3c72d", "content_text": "> Amazon's CEO Doug Herrington explains how conversational AI represents the next major shift in retail, enabling customers to ask questions and receive personalized product recommendations instead of browsing traditional search results. Alexa for Shopping uses agentic AI to help customers navigate hundreds of millions of products through natural conversation.", "date_published": "2026-08-13T15:22:22.715665+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Amazon's AI Shopping Assistant Transforms Retail with Conversational Search", "url": "https://x.com/amazonnews/status/2087637165500960905", "external_url": "https://x.com/amazonnews/status/2087637165500960905" }, { "id": "0673fe74-ee65-4887-a092-338dd068d93b", "content_text": "> A research benchmark evaluating how well large language models can discover new thermally conductive dielectric materials for advanced semiconductor applications, with a leaderboard tracking computational discoveries and synthesis feasibility.", "date_published": "2026-08-12T23:36:25.797093+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Material Discovery Bench: LLM Research Benchmark", "url": "https://discoveredmaterials.com/research/", "external_url": "https://discoveredmaterials.com/research/" }, { "id": "8f51ee92-62b2-463d-b030-48c0cffe45ca", "content_text": "> Grok 4.6 achieves frontier-level AI intelligence with a score of 61 on the Artificial Analysis Intelligence Index, matching GPT-5.6 Sol while offering significantly lower costs and excelling in agentic tasks like customer service and terminal-based work.", "date_published": "2026-08-12T22:55:47.858974+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Grok 4.6 Benchmarks and Cost Efficiency Analysis", "url": "https://artificialanalysis.ai/articles/grok-4-6-benchmarks-and-analysis", "external_url": "https://artificialanalysis.ai/articles/grok-4-6-benchmarks-and-analysis" }, { "id": "4920875b-2a13-4e9a-a7a1-59c254ff37ed", "content_text": "> Cedar is an open-source policy language and evaluation engine designed for authorization and access control decisions in applications and services.\n\nNow is an excellent time to learn more about Cedar.", "date_published": "2026-08-12T20:44:15.409222+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Cedar Policy", "url": "https://cedarpolicy.com/en", "external_url": "https://cedarpolicy.com/en" }, { "id": "0cfd38fe-b525-49a9-93c2-e5ea646bf0cb", "content_text": "> Dogwood is a runtime verification framework designed to monitor and ensure the safe and correct behavior of AI agents during execution.", "date_published": "2026-08-12T20:43:24.474559+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Introducing Dogwood: runtime verification for AI agents", "url": "https://aws.amazon.com/blogs/opensource/introducing-dogwood-runtime-verification-for-ai-agents/", "external_url": "https://aws.amazon.com/blogs/opensource/introducing-dogwood-runtime-verification-for-ai-agents/" }, { "id": "a8c4a079-5eb1-492c-a050-30a59d42c870", "content_text": "> A deep dive into Gorton, an obscure and quirky font that appears everywhere on New York City signage and vintage keyboards, exploring its mysterious origins, unusual letterforms, and surprising prevalence across the city.\n\nExquisite storytelling for font nerds (and everyone else).", "date_published": "2026-08-12T19:37:48.663880+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Hardest Working Font in Manhattan", "url": "https://aresluna.org/the-hardest-working-font-in-manhattan/", "external_url": "https://aresluna.org/the-hardest-working-font-in-manhattan/" }, { "id": "7d4bff38-d782-46af-af38-528cb389824b", "content_text": "> A searchable database of award-winning nonfiction books that aggregates records from major literary prizes like the National Book Awards, Pulitzer Prize, and others, allowing readers, researchers, and librarians to discover prize-recognized books by subject, award, and publisher.\n\nFun way to find 'what to read next'.", "date_published": "2026-08-12T19:33:56.501889+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Book Prize Index", "url": "https://book-prize-index.vercel.app/", "external_url": "https://book-prize-index.vercel.app/" }, { "id": "c3ea96b9-66b1-4dea-b1c2-69c05ded6869", "content_text": "> A large-scale language model from Qwen featuring 512 experts with a mixture-of-experts architecture, designed for text generation tasks and available on Hugging Face.", "date_published": "2026-08-12T19:31:58.560682+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Qwen 3.8 2.4T Mixture of Experts Model", "url": "https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B", "external_url": "https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B" }, { "id": "7dd25285-e8d9-4bd9-a43c-9d548a7b4c90", "content_text": "> DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model available through OpenRouter with pricing of $0.435/$0.87 per 1M tokens and 1M context window support.", "date_published": "2026-08-12T19:31:27.583018+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "DeepSeek V4 Pro 0813 API Pricing & Benchmarks", "url": "https://openrouter.ai/deepseek/deepseek-v4-pro-0813", "external_url": "https://openrouter.ai/deepseek/deepseek-v4-pro-0813" }, { "id": "ef7de84c-15c3-4ab8-9b65-af6f0995b50c", "content_text": "> AI tools are accelerating development velocity without guardrails, causing projects with weak engineering practices to accumulate technical debt at unsustainable rates and collapse into unmaintainable systems that no one understands.", "date_published": "2026-08-12T15:49:01.412051+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "AI is removing the middle class of software engineering", "url": "https://blog.florianherrengt.com/ai-removing-middle-class-software-engineering.html", "external_url": "https://blog.florianherrengt.com/ai-removing-middle-class-software-engineering.html" }, { "id": "cd6a8d69-e655-4100-b149-30ff3229aa73", "content_text": "> Explores the mathematical capabilities and limitations of large language models following their recent breakthroughs in major open problems, analyzing whether they excel particularly at finding counterexamples versus proofs.", "date_published": "2026-08-12T15:48:32.507478+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "What Sort of Maths Are LLMs Good At?", "url": "https://gowers.wordpress.com/2026/08/12/what-sort-of-maths-are-llms-good-at/", "external_url": "https://gowers.wordpress.com/2026/08/12/what-sort-of-maths-are-llms-good-at/" }, { "id": "2ae90d09-bdc3-4c39-aa04-a5f555f7d235", "content_text": "> OpenAI's latest specialized cybersecurity models (Daybreak Red and Blue) are now available on Amazon Bedrock with enterprise security features, data governance, and pricing that matches OpenAI's first-party rates.", "date_published": "2026-08-11T23:29:25.929306+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "OpenAI Models Now Available on Amazon Bedrock", "url": "https://www.aboutamazon.com/news/aws/bedrock-openai-models", "external_url": "https://www.aboutamazon.com/news/aws/bedrock-openai-models" }, { "id": "b44834e1-3d09-461b-bbc9-91ac58ce1552", "content_text": "> OpenAI announced a preview release of the ChatGPT desktop application for Linux, enabling users to access ChatGPT, ChatGPT Work, and Codex directly within their development environments and browser workflows on supported Linux systems.", "date_published": "2026-08-11T23:02:02.199927+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "ChatGPT Desktop App for Linux Now in Preview", "url": "https://x.com/OpenAI/status/2087231350134980830?s=20", "external_url": "https://x.com/OpenAI/status/2087231350134980830?s=20" }, { "id": "7c30bb20-16ba-493b-a259-20e0ef31864c", "content_text": "> This paper investigates whether large language models can introspect on their internal states by injecting known concepts into model activations and measuring how this influences self-reported awareness. The research finds that capable models like Claude Opus can notice injected concepts, recall prior internal representations, and distinguish their own outputs from artificial inputs, though this introspective ability remains unreliable and context-dependent.", "date_published": "2026-08-11T23:01:25.435181+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Emergent Introspective Awareness in Large Language Models", "url": "https://arxiv.org/abs/2601.01828", "external_url": "https://arxiv.org/abs/2601.01828" }, { "id": "554b2536-4d03-4e11-8e1f-792d3a4d1cfd", "content_text": "> A back-of-the-envelope calculation suggests that feedback loops are not currently strong enough to generate a self-sustaining acceleration, though they appear to be strengthening. We conclude by assessing the plausibility and implications of such an acceleration.", "date_published": "2026-08-11T23:01:05.015372+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Economics of Recursive Self-Improvement", "url": "https://elasticity.institute/rsi-paper.pdf", "external_url": "https://elasticity.institute/rsi-paper.pdf" }, { "id": "9c41d71a-cce0-4483-bd42-5bc4f33271b4", "content_text": "> NVIDIA introduces a lightweight open model and routing library that enables faster, more efficient AI agents with greater control over data and workflows across edge devices, PCs, workstations, data centers, and cloud environments.", "date_published": "2026-08-11T22:57:34.825541+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Nemotron 3.5 Lightning and NeMo Switchyard for Agentic AI", "url": "https://blogs.nvidia.com/blog/nemotron-lightning-switchyard-rtx-dgx/", "external_url": "https://blogs.nvidia.com/blog/nemotron-lightning-switchyard-rtx-dgx/" }, { "id": "0f0fb60b-4ef4-4d1a-a2c7-025bdea6e2d6", "content_text": "> Explores the fundamental relationship between data compression and predictive modeling, examining how compression algorithms work as predictors and the implications for AI and LLMs.", "date_published": "2026-08-11T22:57:18.719244+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Compression is prediction", "url": "https://ngrok.com/blog/compression-is-prediction", "external_url": "https://ngrok.com/blog/compression-is-prediction" }, { "id": "d63c59c9-dba4-45ed-bf81-94bddb36fd8b", "content_text": "> Expedia Group upgraded their lodging ranking system using Keras 3, improving their machine learning infrastructure for hotel search results.", "date_published": "2026-08-11T19:56:00.903956+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How Keras 3 Modernized Expedia's Lodging Ranking Stack", "url": "https://medium.com/expedia-group-tech/how-keras-3-helped-modernise-expedia-groups-lodging-ranking-stack-7fec96f052fd", "external_url": "https://medium.com/expedia-group-tech/how-keras-3-helped-modernise-expedia-groups-lodging-ranking-stack-7fec96f052fd" }, { "id": "85ab92af-86a8-4fd3-8f4d-a49430a45571", "content_text": "> Explores how restraint, curation, and subtraction\u2014exemplified by producer Rick Rubin's invisible hand in shaping iconic music\u2014represent the true creative power in an era of algorithmic noise and constant visibility.", "date_published": "2026-08-11T19:46:50.972618+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Creative Power of Invisibility", "url": "https://www.linkedin.com/pulse/creative-power-invisibility-oana-leonte-j1erc/", "external_url": "https://www.linkedin.com/pulse/creative-power-invisibility-oana-leonte-j1erc/" }, { "id": "ee52d0b7-8d3a-4ae7-be08-a7906a5551c8", "content_text": "> A free, open-source Markdown editor for macOS that lets you customize the writing environment with appearance profiles, optional Vim keys, and local file storage without accounts or telemetry.\n\nI love my growing collection of Markdown editors. This one looks fun.", "date_published": "2026-08-11T19:45:05.126725+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Write.md - Customizable Markdown Editor for macOS", "url": "https://writemd.app/", "external_url": "https://writemd.app/" }, { "id": "de6b5cbf-037c-4a89-bf7a-b4299fa07bfc", "content_text": "> An H3 inference engine implementation for Mac computers, providing MiniMax-based AI inference capabilities.", "date_published": "2026-08-11T19:43:38.551252+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "MiniMax H3 Inference Engine for Mac", "url": "https://github.com/antirez/h3.c", "external_url": "https://github.com/antirez/h3.c" }, { "id": "05e5c888-e77e-4fee-a562-cfc4ebd7949f", "content_text": "> Recently the OpenSSH team have received a large number of security\nbug reports, many of which are findings from AI models or made with\nAI assistance. While many AI reports are determined not to have\nsecurity impact when considered in the context of a realistic\nthreat model, we very much welcome these reports, especially when\ncombined with human triage, analysis, test-cases and particularly\nwhen accompanied by proposed fixes.\n\n> We have seen a number of cases where a security bug identified by\nAI tools is subsequently independently discovered by a different\nresearcher. This suggests that adversaries who do not report bugs\nto OSS projects are likely to be able to discover these bugs too.\nGiven this, the OpenSSH team will, for now, be making more frequent\nreleases to get bugfixes into users' hands more quickly rather than\nbatching them until the next planned release.\n\nMakes sense.", "date_published": "2026-08-11T18:43:51.880024+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "OpenSSH 10.5 Release Notes", "url": "https://www.openssh.org/releasenotes.html#10.5", "external_url": "https://www.openssh.org/releasenotes.html#10.5" }, { "id": "00ff9707-e41b-404c-8e47-9da772da0232", "content_text": "> GPU passthrough in macOS virtual machines, covering how Apple Silicon's architecture handles GPU virtualization, the technical challenges involved, and how the Virtualization framework enables GPU resource sharing between host and guest macOS environments.", "date_published": "2026-08-11T17:55:25.971553+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "11\u201316\u00d7 Faster LLM Inference with llama.cpp", "url": "https://github.com/trycua/cua/blob/main/blog/gpu-passthrough-macos-vms.md", "external_url": "https://github.com/trycua/cua/blob/main/blog/gpu-passthrough-macos-vms.md" }, { "id": "50a2b375-10cd-4748-8090-b8d76c74c9cd", "content_text": "> LFM2.5-2.6B is a 2.6 billion parameter language model developed by Liquid AI, available on Hugging Face, featuring instruction-following capabilities with a chat template supporting system prompts and tool use.", "date_published": "2026-08-11T17:13:54.842170+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "LiquidAI/LFM2.5-2.6B \u00b7 Hugging Face", "url": "https://huggingface.co/LiquidAI/LFM2.5-2.6B", "external_url": "https://huggingface.co/LiquidAI/LFM2.5-2.6B" }, { "id": "e4b6aaeb-7a04-4cc5-a513-ca48bff791dd", "content_text": "> Needle 2 is a 45-million-parameter, 14MB open-source language model designed for tool calling, device control, and structured extraction on low-cost edge hardware like microcontrollers, budget phones, and Raspberry Pis. It runs a full session in 28MB of RAM using CQ2-bit compression, achieving competitive performance against much larger small models on mobile device use benchmarks.\n\nFeels like lots is all happening at once with small mobile-friendly models, but this is a space which has been making steady progress for months now. Encouraging.", "date_published": "2026-08-11T01:10:01.897103+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Cactus Needle 2", "url": "https://cactuscompute.com/needle", "external_url": "https://cactuscompute.com/needle" }, { "id": "41b43915-5dd8-48bc-bbe3-bf99bd9d3c90", "content_text": "> An analysis and critique of claims that dynamic programming languages are more token-efficient than static languages for LLM coding agents, examining methodological flaws in the studies behind those claims. The piece argues that conclusions drawn from trivial benchmark problems don't generalize to real-world coding tasks.", "date_published": "2026-08-11T01:09:52.891552+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "What's the best programming language for coding agents?", "url": "https://danluu.com/pl-tokens/", "external_url": "https://danluu.com/pl-tokens/" }, { "id": "61f3f4ee-2cfd-42d3-bdef-2ad4a354f8da", "content_text": "> The parametron, invented by Eiichi Goto in 1954, was a bistable circuit element using ferrite cores that enabled the development of early Japanese computers, including the PC-1. Its invention and application in computing had significant historical impact on computer engineering in Japan, influencing multiple major electronics manufacturers and nurturing a new generation of engineers.\n\nJapan\u2019s first university-built stored-program computer which became the nation's then-fastest in 1958", "date_published": "2026-08-11T01:09:41.436240+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Parametron", "url": "https://ethw.org/Milestones:Parametron,_1954", "external_url": "https://ethw.org/Milestones:Parametron,_1954" }, { "id": "ec4eedb5-4b4f-4904-8e68-1aeed37e7fdb", "content_text": "> An analysis of over 50,000 boat names extracted from NOAA's AIS vessel traffic data, exploring the creative, humorous, and pop-culture-inspired names that boat owners choose, alongside insights into who owns boats in the United States.\n\nAn important dataset, but the all-time best boat name of all time is clearly: Earn Trussed.", "date_published": "2026-08-11T01:04:30.381758+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "50,000 boat names", "url": "https://www.beautifulpublicdata.com/boat-names/", "external_url": "https://www.beautifulpublicdata.com/boat-names/" }, { "id": "788b0328-6d03-4602-9b0b-f5a0d17c7109", "content_text": "> The piece explores how AI adoption momentum within organizations tends to concentrate among early adopters and struggles to spread more broadly, drawing on the example of Benjamin Franklin's Junto club to argue that making the *process* of learning visible \u2014 not just sharing outputs \u2014 is key to organizational change.", "date_published": "2026-08-11T00:16:21.883906+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How This Was Made", "url": "https://mattwood.blog/essays/2026/08/how-this-was-made/", "external_url": "https://mattwood.blog/essays/2026/08/how-this-was-made/" }, { "id": "515eccfe-258f-4c13-84f8-2fe9f98cd3ff", "content_text": "> A critique of the trend of \"humanising\" LLM outputs through prompting techniques, arguing that this approach is the wrong abstraction for addressing verbosity and quirks in AI-generated text.", "date_published": "2026-08-10T23:04:28.283382+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Humanising LLM Outputs is Dumb", "url": "https://kuber.studio/blog/Reflections/Humanising-LLM-Outputs-is-Actually-Dumb", "external_url": "https://kuber.studio/blog/Reflections/Humanising-LLM-Outputs-is-Actually-Dumb" }, { "id": "c054d0ab-6140-4e0f-bd9b-8e9b06af1ad5", "content_text": "> Herdr, an open-source runtime for managing CLI coding agents in terminals, is joining Y Combinator after growing to 25,000 GitHub stars and 340,000 downloads as a solo project. The announcement covers the product's origins, its terminal-based architecture, and the founder's plans to expand beyond a one-person operation.", "date_published": "2026-08-06T21:23:01.707659+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Herdr is joining Y Combinator. The runtime stays open.", "url": "https://herdr.dev/blog/herdr-is-joining-y-combinator/", "external_url": "https://herdr.dev/blog/herdr-is-joining-y-combinator/" }, { "id": "f79e4a2b-1736-4476-9c8d-18640bffe869", "content_text": "> An essay arguing that as AI collapses the cost and effort of building software, the filtering function that effort once provided disappears, leaving \"taste\" \u2014 the judgment of what deserves to exist \u2014 as the remaining meaningful differentiator in software craft.", "date_published": "2026-08-06T21:20:10.613974+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Taste Is All That's Left", "url": "https://notashelf.dev/posts/taste-is-all-thats-left", "external_url": "https://notashelf.dev/posts/taste-is-all-thats-left" }, { "id": "c7d7296a-a3ac-4e2f-980b-68797cae11fc", "content_text": "> AMD's acquisition of AI chip startup Taalas aims to enhance inference performance by physically encoding AI models directly into silicon hardware. The deal represents AMD's effort to compete in the AI accelerator market by leveraging Taalas's approach of optimizing chips at the hardware level for specific AI workloads.", "date_published": "2026-08-06T21:06:45.936868+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "AMD acquires AI chip startup Taalas to boost inference performance by etching models into silicon", "url": "https://www.theregister.com/systems/2026/08/06/amd-acquires-ai-chip-startup-taalas-to-boost-inference-performance-by-etching-models-into-silicon/5284344", "external_url": "https://www.theregister.com/systems/2026/08/06/amd-acquires-ai-chip-startup-taalas-to-boost-inference-performance-by-etching-models-into-silicon/5284344" }, { "id": "5b505b32-1012-4a74-bed0-66e90e6606f3", "content_text": "> OpenAI's work on improving the GPT-4.5 or a related model's performance on solving problems, likely focusing on enhancements to reasoning, accuracy, or problem-solving capabilities within ChatGPT. The content likely details technical improvements, benchmark results, or methodology changes made to advance the model's abilities.", "date_published": "2026-08-06T19:46:04.447721+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Improving Gpt 5 6 Sol In Chatgpt", "url": "https://openai.com/index/improving-gpt-5-6-sol-in-chatgpt/", "external_url": "https://openai.com/index/improving-gpt-5-6-sol-in-chatgpt/" }, { "id": "186d8619-0a44-4f76-9362-107f5bdffdd9", "content_text": "> Discovery Loop is an AI company focused on automating scientific and engineering experimental loops, using frontier AI models and large-scale computational infrastructure to parallelize and accelerate the process of discovery. The company, founded by Jeff Dean, Sanjay Ghemawat, Quoc Le, and Oriol Vinyals, begins with automating machine learning research before expanding to tackle broader scientific grand challenges.", "date_published": "2026-08-05T17:13:16.779387+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Discovery Loop", "url": "https://www.discoveryloop.com/", "external_url": "https://www.discoveryloop.com/" }, { "id": "3fa0663b-ad78-4d07-afc3-46dea50283c8", "content_text": "> Construction of a production-grade agentic harness for LLMs, covering typed tool validation, parallel execution via dependency graphs, multi-tier memory, verification hierarchies, role separation (Planner/Worker/Critic), and budget controls. A city comparison agent serves as the running example to illustrate how these primitives compose into a reliable, debuggable system.", "date_published": "2026-08-05T15:41:30.117305+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Building an Advanced Agentic Harness", "url": "https://data4sci.com/blog/building-an-advanced-agentic-harness", "external_url": "https://data4sci.com/blog/building-an-advanced-agentic-harness" }, { "id": "b22e8d01-b0ba-4a75-9231-5aca8781933c", "content_text": "> TL;DR: Scientific invention requires manipulative abduction and physical simulation\n", "date_published": "2026-08-05T15:40:06.516756+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "LLMs can't jump", "url": "https://openreview.net/forum?id=klU4737opt", "external_url": "https://openreview.net/forum?id=klU4737opt" }, { "id": "2ba86b41-6a57-4209-b940-71aefcf532d9", "content_text": "> An argument that intelligence and technological capability are often not the primary barriers to real-world progress, using examples like medical regulation and housing policy to illustrate how political will, institutional constraints, and regulatory bottlenecks frequently matter more than raw intelligence or AI advancement.", "date_published": "2026-08-05T15:38:14.998420+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Intelligence is not the main bottleneck", "url": "https://www.writingruxandrabio.com/p/intelligence-is-not-the-main-bottleneck", "external_url": "https://www.writingruxandrabio.com/p/intelligence-is-not-the-main-bottleneck" }, { "id": "dd9ca296-ef1b-4e0c-a4e3-b4a65ceb8ba9", "content_text": "> A technical article published in ACM Queue likely covering a specific topic in computer science, software engineering, or systems design, as is typical of the publication's focus on practical and research-oriented computing topics.\n\nThe myths are:\n\n1. Developers Spend Most of Their Time Writing Code\n2. Writing Code Is the Bottleneck\n3. Lines of Code Written by AI Is the Best Measure of Impact\n4. AI Helps All Tasks and Engineers Equally\n5. AI Will Turn Individual Developers into 10x Developers\n6. It\u2019s Up to Each Developer to Make AI Work\n7. High-Performing AI Tools Will Be Adopted Automatically\n8. With GenAI, Enterprises Can Innovate at Startup Speed\n", "date_published": "2026-08-05T01:04:10.614513+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "8 Myths of Software Development with AI", "url": "https://queue.acm.org/detail.cfm?id=3807963", "external_url": "https://queue.acm.org/detail.cfm?id=3807963" }, { "id": "5fe58159-195e-48a9-af74-5b6fbc1a65ac", "content_text": "> A discussion of Pi, a minimalist AI coding harness with only 4 tools and under 1,000 tokens in its system prompt, which achieves industry-leading performance at lower cost by keeping context lean and avoiding excessive orchestration layers. Case studies from Databricks and Shopify illustrate how Pi's minimal design outperforms more complex coding agents on real-world tasks.\n\nPi is the coding harness that chooses minimalism on purpose. It comes out of the box with only 4 tools, and its system prompt and tool definitions come in below 1,000 tokens. The idea being that most work can be done with the basics, and if you want more, build it.", "date_published": "2026-08-05T01:02:47.442524+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Pi, Minimal and Performant", "url": "https://earendil.com/posts/pi-autoresearch-and-databricks/", "external_url": "https://earendil.com/posts/pi-autoresearch-and-databricks/" }, { "id": "39ec3a60-2f1b-4d2f-a7b7-f85af0e832b9", "content_text": "> An analysis of how previous technology revolutions (PCs, the internet, smartphones) typically increased workloads and benefited companies more than employees, and why AI may be different by actually performing work rather than just accelerating it. The piece argues that while AI is currently causing workplace disruption, it has the potential to free up time for creative thinking and collaboration rather than simply raising productivity expectations.", "date_published": "2026-08-04T23:00:00.287067+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Most tech revolutions made work worse for employees. AI could be the exception: this+that", "url": "https://www.thisandthat.chat/blog/most-tech-revolutions-made-work-worse-for-employees/", "external_url": "https://www.thisandthat.chat/blog/most-tech-revolutions-made-work-worse-for-employees/" }, { "id": "ae594c3a-86ba-4949-b143-830aa9662a41", "content_text": "> An exploration of how humans solve Coverage Path Planning problems, using an interactive lawn-mowing experiment where tens of thousands of participants found near-optimal paths, compared to the mathematical challenge these routing problems pose for computers. The piece examines why humans are surprisingly efficient at spatial navigation tasks like mowing or vacuuming despite the astronomical number of possible route combinations.\n\nFrom the article, a good description of algorithms and heuristics.\n\n> A slight detour, if we may. There are two main ways to tackle a problem like this. Simplifying things a bit, there are algorithms which guarantee the optimal path, and heuristics which use shortcuts to find a \u201cgood enough\u201d path, fast.\n\nAlso, this domain is a keeper.\n\n", "date_published": "2026-08-04T22:13:09.160469+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Why some people mow a lawn better than others", "url": "https://pudding.cool/2026/06/mow/", "external_url": "https://pudding.cool/2026/06/mow/" }, { "id": "80231a7b-feb1-46a9-b953-02450095a0af", "content_text": "> Kiro Crew is an open-source, persistent AI development workspace that retains memory and context across sessions, learns from user workflows, and coordinates autonomous agents to handle tasks like issue triage, CI/CD monitoring, and scheduled jobs \u2014 even when the user is away. It includes multi-layered security, a knowledge graph with vector search, and editable lessons and skills stored as Markdown files.", "date_published": "2026-08-04T18:15:41.669253+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Kiro Crew", "url": "https://kiro.dev/crew/", "external_url": "https://kiro.dev/crew/" }, { "id": "9a5b38d2-2e34-480d-ae6d-cf69aee7091d", "content_text": "Steve Yegge discusses AI coding agent techniques, particularly using \"loops and graphs\" harnesses to tackle large problems autonomously, and introduces his custom harness called Wheelhouse built for his long-running MMORPG project Wyvern. He argues that reusable harness frameworks are a dead end and that effective AI harnesses must be bespoke and tightly integrated into the specific application being built.\n\n> Don't be special, stay out in front, and you will see the future clear as day. \n\nGood advice.", "date_published": "2026-08-04T18:11:57.802890+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Shape of Things to Come, Part 1: The Continuous Thunderdome", "url": "https://yegge.ai/essays/the-shape-of-things-to-come/", "external_url": "https://yegge.ai/essays/the-shape-of-things-to-come/" }, { "id": "dd4a67f1-ec45-4191-a343-3da49e02a1a5", "content_text": "> TerminalWidget is a macOS, iOS, and iPadOS app that allows users to display terminal command output, scripts, API data, and Shortcuts directly in native widgets, with support for rich text formatting, progress bars, charts, sparklines, and images. It includes a full CLI, AppleScript support, URL scheme automation, and syncs across Apple devices via iCloud.\n\nGreat way to enable your agent to share status, notifications, data, etc. Cool.", "date_published": "2026-08-04T18:03:47.111066+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "TerminalWidget for Mac, iPhone, and iPad", "url": "https://terminalwidget.app/", "external_url": "https://terminalwidget.app/" }, { "id": "9570f009-87bd-4613-94a3-694877217885", "content_text": "> Marin is an open collaborative lab for building foundation models from scratch, sharing all code, data, experiments, and results transparently in real-time. It invites open-source contributors to participate in model training, architecture research, and experiments, with publicly documented models like Marin-8B and Marin-32B that compete with leading open-weight models.", "date_published": "2026-08-04T01:32:51.503697+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Marin", "url": "https://marin.community/", "external_url": "https://marin.community/" }, { "id": "5fcfac5c-f78a-47c4-b9ac-970f53435ca2", "content_text": "> Open Athena is a nonprofit organization that partners with academic institutions to develop open-source AI foundation models by providing engineering talent, compute resources, and coordination. Current projects include large language models with Stanford, plant genomics DNA models with Cornell, and protein generation models with MIT.", "date_published": "2026-08-04T01:31:52.014559+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Open Athena", "url": "https://openathena.ai/", "external_url": "https://openathena.ai/" }, { "id": "197c5d43-cef0-4e5c-a6c9-62c0590664be", "content_text": "", "date_published": "2026-08-03T20:52:56.299722+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Leiden Declaration on Artificial Intelligence and Mathematics", "url": "https://leidendeclaration.ai/", "external_url": "https://leidendeclaration.ai/" }, { "id": "ba6ffe6d-b758-48af-827c-532923730ed9", "content_text": "A multiplayer agent harness for work. In Slack and on the web.", "date_published": "2026-07-31T23:54:39.839148+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "qm", "url": "https://github.com/yc-software/qm", "external_url": "https://github.com/yc-software/qm" }, { "id": "fa45a3ef-3d68-4408-97b8-9971053cf387", "content_text": "> With this release MCP becomes a stateless protocol that scales on ordinary HTTP infrastructure. Every request now carries protocol version, client info, and client capabilities inside its _meta parameter, eliminating the need for a one-time initialization handshake. Clients that need to learn what a server supports can call the new server/discover method at any point.", "date_published": "2026-07-28T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How AgentCore Gateway Supports the MCP 2026-07-28 Spec", "url": "https://aws.amazon.com/blogs/machine-learning/how-agentcore-gateway-supports-the-mcp-2026-07-28-spec/", "external_url": "https://aws.amazon.com/blogs/machine-learning/how-agentcore-gateway-supports-the-mcp-2026-07-28-spec/" }, { "id": "7693e704-78b9-4148-8fe0-1466036ea8b1", "content_text": "> Companies that invest heavily in AI grow headcount 10% over the two years following adoption. Entry-level headcount grows 12%.", "date_published": "2026-07-23T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Does AI eliminate jobs? Economists find heavy adopters hire more.", "url": "https://ramp.com/data/ai-jobs-impact", "external_url": "https://ramp.com/data/ai-jobs-impact" }, { "id": "6d85750b-32ad-430d-8c95-428a527e9803", "content_text": "> Jensen Huang positioning: \"Fireworks is the TSMC of AI Factories...\"", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Fireworks AI: Specialized Intelligence Infrastructure", "url": "https://fireworks.ai/", "external_url": "https://fireworks.ai/" }, { "id": "6aca8003-7d13-422b-8ce9-4fe369045f41", "content_text": "", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Strands AI Functions: Python Library for Verified Agent Workflows", "url": "https://github.com/strands-labs/ai-functions/", "external_url": "https://github.com/strands-labs/ai-functions/" }, { "id": "ab7db5af-6811-456b-99d7-4a51bb8eefab", "content_text": "> In both cases, the country is celebrating a major birthday in the midst of a rising stock market and widespread fears of \"technological unemployment\" (mechanical power then vs. AI now); giddy wealth is coiled with economic anxiety; technology has transformed the way that people get information, mind-wiring us to a global cacophony of far-flung emotions (radio then vs. social media now).", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "America, 1926: What a Forgotten 100-Year-Old Report Says About Who We Are", "url": "https://www.derekthompson.org/p/america-1926-an-absurdly-deep-dive", "external_url": "https://www.derekthompson.org/p/america-1926-an-absurdly-deep-dive" }, { "id": "dd4f17f4-b7e6-420a-bfdd-2b8dc2f384b7", "content_text": "> From the homepage: \"Build something Lovable. Create apps and websites by chatting with AI.\" Templates include spatial canvas tools, blogs, habit trackers, code-powered presentation builders, and e-commerce stores. Key metrics highlighted: millions of projects built, with substantial new projects created per week.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Lovable: AI App Builder", "url": "https://lovable.dev/", "external_url": "https://lovable.dev/" }, { "id": "bf0c03a6-e218-46b2-bf52-1c3dbac9e0a7", "content_text": "> Transfer finding: \"Learn it on the wrist. Use it anywhere on the body. Pretrain once on the wrist, then point the model anywhere. It holds up on body placements, and even sensor types like gyroscope and magnetometer, that it never saw during training.\"", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Inertia-1: Unified Motion Foundation Model from Wearable Sensors", "url": "https://yang-ai-lab.github.io/Inertia-1/", "external_url": "https://yang-ai-lab.github.io/Inertia-1/" }, { "id": "7eec841a-b846-4109-8f72-ec2904440909", "content_text": "> Post-conditions pattern:\n```python\n@ai_function(post_conditions=[check_length, check_style], max_attempts=5)\ndef summarize_meeting(transcripts: str) -> MeetingSummary:\n \"\"\"Write a summary of the following meeting in less than 50 words.\"\"\"\n```\nPost-conditions can be plain Python assertions or other AI Functions. The function only returns once eve", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Strands AI Functions: Post-Conditions and Multi-Agent Teams as Python Functions", "url": "https://github.com/strands-labs/ai-functions", "external_url": "https://github.com/strands-labs/ai-functions" }, { "id": "d9771e85-2c60-4a8e-a4fd-c50b1fd9e009", "content_text": "> Custom agents provide a way to customize Kiro behavior by defining specific configurations for different use cases. Each custom agent is defined by a configuration file that specifies which tools the agent can access, what permissions it has, and what context it should include.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Kiro CLI Custom Agents", "url": "https://kiro.dev/docs/cli/custom-agents/", "external_url": "https://kiro.dev/docs/cli/custom-agents/" }, { "id": "900b4721-f47e-49c9-97b4-2759f8e8af2c", "content_text": "> What the techno-determinism angle misses is: Why did these technologies catch on in the first place? Not every technology people have invented has caught on the way these forms have. They caught on in large part because of this impulse people have to live in a uni-context.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Uni-Context: A Philosopher's One-Word Theory to Explain Why the World Feels So Weird", "url": "https://www.derekthompson.org/p/a-philosophers-one-word-theory-to", "external_url": "https://www.derekthompson.org/p/a-philosophers-one-word-theory-to" }, { "id": "8ba1e900-65ad-4afb-9b63-1a925716f2d9", "content_text": "> Agents need clarity above everything else \u2014 APIs where reading the code tells you exactly what it does.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Designing APIs for Agents - Freestyle", "url": "https://www.freestyle.sh/blog/opinion/designing-apis-for-agents", "external_url": "https://www.freestyle.sh/blog/opinion/designing-apis-for-agents" }, { "id": "7d8a04ea-9874-4527-815d-54658dd9f511", "content_text": "> Maintainers are now facing an assault on two fronts. The barrier to entry for generating code has dropped to zero... flooded repository gates with an alarming volume of low-quality, AI-generated pull requests. Maintainers who once spent their time writing code are now forced to become full-time, unpaid code reviewers.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Zero-Cost Fallacy: Open Source in the Agentic Era (Thoughtworks)", "url": "https://www.thoughtworks.com/insights/blog/open-source/zero-cost-fallacy-open-source-agentic-era", "external_url": "https://www.thoughtworks.com/insights/blog/open-source/zero-cost-fallacy-open-source-agentic-era" }, { "id": "b148cb3d-8328-40ff-8e5b-40023f1aa01f", "content_text": "> Key positioning: \"Stop relying on generic AI models. Databricks has the tools to build agent systems that deliver accurate, data-driven results.\"", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Databricks AI: Agent Bricks and Unity AI Gateway", "url": "https://www.databricks.com/product/artificial-intelligence", "external_url": "https://www.databricks.com/product/artificial-intelligence" }, { "id": "a3b65f14-f00d-49aa-b68e-1f92d8640a76", "content_text": "> In the run that used GPT-5.5 for both planners and workers, the workers alone cost $9,373. In the run where Opus 4.8 did the planning and Composer 2.5 did the work, the entire worker fleet cost $411.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Agent Swarms and the New Model Economics", "url": "https://cursor.com/blog/agent-swarm-model-economics", "external_url": "https://cursor.com/blog/agent-swarm-model-economics" }, { "id": "8567cd81-0d8b-4505-b40b-35dab53b54fd", "content_text": "> Valkey is an open source (BSD) high-performance key/value datastore that supports a variety of workloads such as caching, message queues, and can act as a primary database. The project is backed by the Linux Foundation, ensuring it will remain open source forever.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Valkey: Open Source High-Performance Key/Value Datastore", "url": "https://valkey.io/", "external_url": "https://valkey.io/" }, { "id": "c5a78c19-fd25-4614-a165-8a1b1ef5179d", "content_text": "> ARC-AGI-3 gives an agent a game environment, without an explanation of what it is seeing. At each step, the agent receives a 64\u00d764 grid of 16 color indices and a set of legal actions. The environment supplies no object list, rule sheet, stated goal, or shaped reward.", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Schema: Frontier Models with the Right Harness Achieve ~99% on ARC-AGI-3 Public", "url": "https://schema-harness.github.io/", "external_url": "https://schema-harness.github.io/" }, { "id": "ccdd7f21-48d8-48fe-8b17-e98c352db069", "content_text": "> Key research highlights at ICML 2026:\n- FlashAttention-4: Algorithm and kernel pipelining co-design for asymmetric hardware scaling (Tri Dao et al.)\n- Mamba-3: Next-generation state space model (CMU + Princeton + Together + Cartesia)\n- DeepSWE: Fully open-sourced state-of-the-art coding agent trained by scaling RL\n- Cache-aware prefill-decode disag", "date_published": "2026-07-20T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Together AI: Research-Driven Inference and Training Platform", "url": "https://www.together.ai/", "external_url": "https://www.together.ai/" }, { "id": "2b0673c6-ab15-4c2d-8bc1-15136afe01df", "content_text": "", "date_published": "2026-07-15T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Anti-Mac Interface", "url": "https://www.nngroup.com/articles/anti-mac-interface/", "external_url": "https://www.nngroup.com/articles/anti-mac-interface/" }, { "id": "692eac94-2d90-4e88-a96d-c85e250e0035", "content_text": "> A metastable failure is a self-sustaining congestive collapse in which a system degrades in response to a transient stressor (e.g., a load surge) but fails to recover after the stressor is removed. These rare but potentially catastrophic events are notoriously hard to diagnose and mitigate, sometimes causing prolonged outages affecting millions of ", "date_published": "2026-07-15T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Analyzing Metastable Failures", "url": "https://www.amazon.science/publications/analyzing-metastable-failures", "external_url": "https://www.amazon.science/publications/analyzing-metastable-failures" }, { "id": "3032f045-2392-471f-8c89-c163dd2ec25c", "content_text": "> Model specs:\n- 975B total params, 41B active (Mixture-of-Experts)\n- Inkling-Small: 12B active params\n- 1M token context window\n- Pretrained on 45T tokens (text, images, audio, video)\n- Controllable thinking effort (0.2 to 0.99 sweep)", "date_published": "2026-07-15T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Inkling: Thinking Machines' Open-Weights Model (and Tinker Fine-Tuning Platform)", "url": "https://thinkingmachines.ai/news/introducing-inkling/", "external_url": "https://thinkingmachines.ai/news/introducing-inkling/" }, { "id": "51f5b205-c3a8-446a-9987-462fc2653fc7", "content_text": "> There are only two things you can say with certainty about token prices: we're in a supply crunch, and this is unstable. All of the variables are in play, and the market will get shaken out over the next few years to arrive at a new equilibrium.", "date_published": "2026-07-09T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Ways to think about token pricing", "url": "https://www.ben-evans.com/benedictevans/2026/7/9/ways-to-think-about-token-pricing", "external_url": "https://www.ben-evans.com/benedictevans/2026/7/9/ways-to-think-about-token-pricing" }, { "id": "42eb2859-0d65-48c5-a08b-3a39c4f005f2", "content_text": "> We formalize this bottleneck as the intent-execution gap: the mismatch between what the model intends and what the harness executes, and vice versa. For example, in trying to revise code, a model may intend to edit a single instance of a function, while the harness accidentally modifies multiple instances.", "date_published": "2026-07-08T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Bridging Intent and Execution in Agentic Systems", "url": "https://www.amazon.science/blog/bridging-intent-and-execution-in-agentic-systems", "external_url": "https://www.amazon.science/blog/bridging-intent-and-execution-in-agentic-systems" }, { "id": "0a0b793b-792e-4f20-8802-6e5d82da546e", "content_text": "> *Note: LinkedIn blocked direct crawling of this article. The following is reconstructed from the article title, author context, and surrounding discourse on the same topic.*", "date_published": "2026-07-07T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The New Network is You", "url": "https://www.linkedin.com/pulse/new-network-you-merritt-robinson--6l68c?utm_source=share&utm_medium=member_android&utm_campaign=share_via", "external_url": "https://www.linkedin.com/pulse/new-network-you-merritt-robinson--6l68c?utm_source=share&utm_medium=member_android&utm_campaign=share_via" }, { "id": "193e6d70-11cb-4eb7-ba1f-f83df1702e67", "content_text": "", "date_published": "2026-07-06T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Neuronpedia Jacobian Lens: Interactive J-Space Explorer for Qwen3.6-27B", "url": "https://www.neuronpedia.org/qwen3.6-27b/jlens", "external_url": "https://www.neuronpedia.org/qwen3.6-27b/jlens" }, { "id": "e7c6b790-fc93-4e94-9a70-56458bcf6fa4", "content_text": "> Sponsorship: Diamond level, Booth B207, Seoul South Korea", "date_published": "2026-07-05T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Amazon at ICML 2026: Research Presence and Booth Schedule", "url": "https://www.linkedin.com/pulse/amazon-icml-2026-papers-imen-grida-ben-yahia-ph-d--znhse", "external_url": "https://www.linkedin.com/pulse/amazon-icml-2026-papers-imen-grida-ben-yahia-ph-d--znhse" }, { "id": "974e9370-89a8-4d75-a7cc-0a5f05141b4e", "content_text": "> Core thesis: \"The phrase 'frontier model' is starting to mean two things. One is a checkpoint. The other is a system boundary.\"", "date_published": "2026-06-29T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Micro-Agent: Beat Frontier Models with Collaboration inside Model API", "url": "https://vllm.ai/blog/2026-06-29-micro-agent-frontier-models", "external_url": "https://vllm.ai/blog/2026-06-29-micro-agent-frontier-models" }, { "id": "9f91f343-c679-457c-941c-888fa82f0b62", "content_text": "> A Distilled Knowledge Skill (DKS): an entire domain (Strands Agents, Amazon Bedrock, Bedrock AgentCore) reduced to its executable essence, and kept current as the surface shifts month to month. The research is already done; you skip straight to building.", "date_published": "2026-06-27T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "AWS Bedrock AgentCore Skill - Claude Code Plugin", "url": "https://github.com/ferdinandobons/AWSBedrockAgentCoreSkill", "external_url": "https://github.com/ferdinandobons/AWSBedrockAgentCoreSkill" }, { "id": "e76bc075-b6b6-455c-979b-01af76eebf31", "content_text": "> Over the past few months, four AI giants quietly rebuilt the same thing at once: AWS, Microsoft, Google, and Anthropic each shipped agent runtime updates that point to the same architectural shift. [...] It is a move from request-level load balancing to session-aware execution.", "date_published": "2026-06-27T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Agent Session-Aware Runtime", "url": "https://thenewstack.io/agent-session-aware-runtime/", "external_url": "https://thenewstack.io/agent-session-aware-runtime/" }, { "id": "a209183b-395f-42c0-94bc-1b6356f9690f", "content_text": "> Architecture: Combines a multinomial diffusion model for ingredient selection with a score-based generative model for ingredient quantification, together generating complete burger recipes defined by 146 ingredients and their quantities.", "date_published": "2026-06-26T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Generative artificial intelligence creates delicious, sustainable, and nutritious burgers", "url": "https://www.nature.com/articles/s41538-026-00953-x", "external_url": "https://www.nature.com/articles/s41538-026-00953-x" }, { "id": "a2d40910-204b-4beb-9604-65bb05d799f8", "content_text": "> `gog mcp` runs a typed MCP server over stdio for agent clients that need a permissioned Google Workspace tool surface. It intentionally does not expose a generic shell/argv bridge. Each MCP tool has a fixed schema and maps to a specific gog operation. MCP defaults are read-only. Write tools are hidden unless the server is started with `--allow-wri", "date_published": "2026-06-26T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "gogcli spec: Unified Go CLI for Google Workspace", "url": "https://gogcli.sh/spec.html", "external_url": "https://gogcli.sh/spec.html" }, { "id": "d7a455ee-d12c-4f36-933e-25f6e1042c41", "content_text": "> His ZX Spectrum computers brought affordable personal computing to the masses and sold in their millions across the world. But his attempt to launch an electric vehicle was not successful, and caused him severe financial problems.", "date_published": "2026-06-14T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Sir Clive Sinclair: Tireless Inventor Ahead of His Time", "url": "https://www.bbc.com/news/science-environment-29985976", "external_url": "https://www.bbc.com/news/science-environment-29985976" }, { "id": "6d19e5a5-1568-4e7e-b826-8bb4c9a4afb9", "content_text": "> It's not that agents can on their own construct arbitrarily challenging proofs. But models are enormously helpful, and broaden the set of people who can use these tools productively. With formal methods being easier to use than ever, it's worth reconsidering the old cost/benefit calculus.", "date_published": "2026-06-14T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Formal Methods at Jane Street: Agentic Coding Changes the Calculus", "url": "https://blog.janestreet.com/formal-methods-at-jane-street-index/?from_theconsensus=1", "external_url": "https://blog.janestreet.com/formal-methods-at-jane-street-index/?from_theconsensus=1" }, { "id": "07309017-7d0f-4e71-8d33-9e29d243b8dd", "content_text": "", "date_published": "2026-06-12T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling", "url": "https://arxiv.org/abs/2606.13473", "external_url": "https://arxiv.org/abs/2606.13473" }, { "id": "22591151-d114-4c39-9997-993df8d2c95d", "content_text": "", "date_published": "2026-06-12T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "General-purpose large language models outperform specialized clinical AI tools on medical benchmarks", "url": "https://www.nature.com/articles/s41591-026-04431-5", "external_url": "https://www.nature.com/articles/s41591-026-04431-5" }, { "id": "ff88fee8-ff0a-42d9-b055-5617cbd161e1", "content_text": "> SIA operates by coordinating three main types of AI agents that work together to continuously improve task performance: - Meta-Agent: Reads the task description and generates an initial Target Agent tailored to the task. - Target/Task Specific Agent: Attempts to complete the task and records its actions and results. - Feedback/Improvement Agent: Re", "date_published": "2026-06-11T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "SIA: Self Improving AI Framework", "url": "https://github.com/hexo-ai/sia", "external_url": "https://github.com/hexo-ai/sia" }, { "id": "783f0fb9-15b6-4db1-ac29-67fe8c999070", "content_text": "> They attributed the AI-enabled gain to three factors multiplying together: acceleration of low-judgment work (1.5x), higher focus on high-judgment work with no context-switching (1.5x), and instant access to agent-captured domain expertise (1.5x). Remove any one factor and the gains collapse.", "date_published": "2026-06-10T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How Frontier Teams Are Reinventing AI-Native Development", "url": "https://aws.amazon.com/blogs/machine-learning/how-frontier-teams-are-reinventing-ai-native-development/", "external_url": "https://aws.amazon.com/blogs/machine-learning/how-frontier-teams-are-reinventing-ai-native-development/" }, { "id": "bb3149b7-a319-4587-8fd7-fa5d80c83e67", "content_text": "> The LinkedIn post (content not directly crawlable) is Brad Porter commenting \"We saw this with robotics at Amazon too\" on a shared post about technology deployment gaps. Based on Porter's extensive public commentary:", "date_published": "2026-06-09T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Brad Porter: We Saw This With Robotics at Amazon Too", "url": "https://www.linkedin.com/posts/brad-porter-cobot_we-saw-this-with-robotics-at-amazon-too-share-7470168784201273346-XpL8/", "external_url": "https://www.linkedin.com/posts/brad-porter-cobot_we-saw-this-with-robotics-at-amazon-too-share-7470168784201273346-XpL8/" }, { "id": "1e722418-ef36-415f-8212-3d09d034ff56", "content_text": "> Key argument (from David, 1990, American Economic Review, Vol. 80, No. 2, pp. 355-361):", "date_published": "2026-06-09T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Dynamo and the Computer: An Historical Perspective on the Modern Productivity Paradox", "url": "https://www.almendron.com/tribuna/wp-content/uploads/2018/03/the-dynamo-and-the-computer-an-historical-perspective-on-the-modern-productivity-paradox.pdf", "external_url": "https://www.almendron.com/tribuna/wp-content/uploads/2018/03/the-dynamo-and-the-computer-an-historical-perspective-on-the-modern-productivity-paradox.pdf" }, { "id": "97ea8b21-549c-4bef-8773-6fbdb173f228", "content_text": "> Loop engineering is replacing yourself as the person who prompts the agent. You design the system that does it instead. A loop here can be thought of a recursive goal where you define a purpose and the AI iterates until complete. It's roughly five building blocks and Claude Code and Codex both have all five now.", "date_published": "2026-06-09T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Loop Engineering - by Addy Osmani - Elevate", "url": "https://addyo.substack.com/p/loop-engineering", "external_url": "https://addyo.substack.com/p/loop-engineering" }, { "id": "8691c6a4-3c7d-4b07-9123-c790189da140", "content_text": "> \"There are three phases to AI coding. The first is autocomplete. The second is AI-assisted chat panels in IDEs. We are now entering the Agent-First era.\" \u2014 Zach Lloyd", "date_published": "2026-06-04T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Lessons I've Learned at Warp Building for Developers", "url": "https://www.linkedin.com/pulse/lessons-ive-learned-warp-building-developers-zach-lloyd-cbgoc", "external_url": "https://www.linkedin.com/pulse/lessons-ive-learned-warp-building-developers-zach-lloyd-cbgoc" }, { "id": "962a7ba4-1458-4836-b456-b30287bda4a8", "content_text": "> Issue 1: Pod stuck in Pending state - Agent correctly identified missing Fargate profile for the test namespace.", "date_published": "2026-06-03T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Exploring AWS DevOps Agent Part 3 - Troubleshooting Issues", "url": "https://awstip.com/exploring-aws-devops-agent-part-3-1d16b7f2dd72", "external_url": "https://awstip.com/exploring-aws-devops-agent-part-3-1d16b7f2dd72" }, { "id": "9358b96b-9366-41ae-b0e2-ef9048a72354", "content_text": "> Key details from coverage (NYT article paywalled; sourced from aggregated reporting):", "date_published": "2026-06-02T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How One Tech Company Created 13 New Types of Jobs Because of A.I.", "url": "https://www.nytimes.com/2026/06/01/technology/box-13-new-types-jobs-ai.html", "external_url": "https://www.nytimes.com/2026/06/01/technology/box-13-new-types-jobs-ai.html" }, { "id": "dd542040-d097-46ad-8017-37ccd8ab2485", "content_text": "> AI Doomerism = a Secular Apocalypse Narrative. Practically, the \"AI doom gospel\" translates traditional religious forms into technical language. In the rationalist/effective altruist discourse, God becomes a superintelligence, prophecy becomes timelines and probability estimates, and hell becomes human extinction from AI.", "date_published": "2026-06-02T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "First They Built a Secular Apocalypse Belief System. Now They Want Religious Authority.", "url": "https://www.aipanic.news/p/first-they-built-a-secular-apocalypse", "external_url": "https://www.aipanic.news/p/first-they-built-a-secular-apocalypse" }, { "id": "12831b03-6724-45fb-aa04-ee779dbf1c3c", "content_text": "> I've been spending a lot of time thinking about the shape of the capabilities of coding agents. What they're good at now, what they're going to be good at. What they're bad at now, how much of that is inherent and how much is transient.", "date_published": "2026-06-02T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "What's Easy Now? What's Hard Now? - Marc's Blog", "url": "https://brooker.co.za/blog/2026/05/18/whats-easy-whats-hard.html", "external_url": "https://brooker.co.za/blog/2026/05/18/whats-easy-whats-hard.html" }, { "id": "8e0f8185-34f1-4000-ae11-f42a7ed9e504", "content_text": "> Problem: Adapting foundation models to a morphologically rich language (Azerbaijani) with limited training data, no existing blueprint for efficient LLM training in that language.", "date_published": "2026-06-01T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Training Azerbaijani Language Models on Amazon SageMaker AI", "url": "https://www.linkedin.com/posts/sabirmardan_amazon-sagemaker-amazon-share-7467195020136144897-lSzD/?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/sabirmardan_amazon-sagemaker-amazon-share-7467195020136144897-lSzD/?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "d46755b3-e3fb-4239-b0ae-e2e972892507", "content_text": "> \"Learning can only take place through the attempt to solve a problem and therefore only takes place during activity.\"", "date_published": "2026-06-01T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Economic Implications of Learning by Doing - Kenneth Arrow (1962)", "url": "https://www.haverford.edu/sites/default/files/Arrow1962.pdf", "external_url": "https://www.haverford.edu/sites/default/files/Arrow1962.pdf" }, { "id": "99c9907a-2f15-4aa5-a23f-a513f7ba8e8f", "content_text": "> The 1997 analogy: We are in an era of radical uncertainty where the technology is transformative but most high-value use cases haven't been built yet and the \"winners\" are not yet clear. Just as it was impossible in 1997 to predict that a search engine with a quirky logo would reshape the world, we cannot yet see the final shape of the AI-driven ec", "date_published": "2026-06-01T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "A Rational Conversation on Where AI Is Actually Going | Benedict Evans on Lenny's Podcast", "url": "https://www.linkedin.com/posts/lennyrachitsky_my-biggest-takeaways-from-benedict-evans-share-7467220148777656320-CvQS/?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/lennyrachitsky_my-biggest-takeaways-from-benedict-evans-share-7467220148777656320-CvQS/?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "c42cead4-2bbd-4a26-b28b-c97d63afdb7b", "content_text": "> Authors: Tianyi Zhou, Dongrui Liu, Leitao Yuan, Jing Shao, Xia Hu", "date_published": "2026-06-01T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "COLLEAGUE.SKILL: Automated AI Skill Generation via Expert Knowledge Distillation", "url": "https://arxiv.org/abs/2605.31264", "external_url": "https://arxiv.org/abs/2605.31264" }, { "id": "0c3e68d4-cbcd-4ef9-a3f7-64f21c23970a", "content_text": "> Mollick's post (May 30, 2026):\n\"It does seem like meaningfully better AI releases are accelerating, especially from OpenAI & Anthropic. To illustrate, I caused this timeline to be created. It only lists new models that scored 3 points or higher over previous models in the Artificial Analysis index.\"", "date_published": "2026-05-31T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Ethan Mollick: Meaningfully Better AI Releases Are Accelerating", "url": "https://www.linkedin.com/posts/emollick_it-does-seem-like-meaningfully-better-ai-share-7466647737384742912-WpIl/", "external_url": "https://www.linkedin.com/posts/emollick_it-does-seem-like-meaningfully-better-ai-share-7466647737384742912-WpIl/" }, { "id": "6a1cba95-ede7-4184-a4da-63601f722122", "content_text": "> The problem: At Meta, significant lines of code per human-landed diff grew by 105.9% year over year and per-developer diff volume rose 51%, with agentic AI responsible for over 80% of that growth. Meanwhile, the share of diffs receiving timely review has declined, exposing a widening gap between code supply and reviewer bandwidth.", "date_published": "2026-05-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Automating Low-Risk Code Review at Meta: RADAR, Risk Calibration, and Review Efficiency", "url": "https://arxiv.org/abs/2605.30208", "external_url": "https://arxiv.org/abs/2605.30208" }, { "id": "2ade86b6-71e9-4687-bf08-9619cc38dfeb", "content_text": "> Thinking Display: See how the agent works through a problem as it happens. Thinking display streams the model's reasoning in real time, so you can follow its logic, catch a wrong turn early, and understand why it chose an approach. Enabled by default; toggle from /settings > Display > Show thinking.", "date_published": "2026-05-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Kiro CLI 2.5: Thinking Display, Subagent Review Loops, and Display Controls", "url": "https://kiro.dev/changelog/cli/2-5/", "external_url": "https://kiro.dev/changelog/cli/2-5/" }, { "id": "5fccd6e5-a33b-4235-bf20-d678930f92cd", "content_text": "> Architecture: Built with Electron, TypeScript, and AWS SDK. Uses Kiro CLI's Agent Client Protocol for conversational AI. Three agents collaborate to build features and improve code/UX quality.", "date_published": "2026-05-29T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "ARchitect: Automated Reasoning Policy Formalization for Bedrock Guardrails", "url": "https://github.com/aws-samples/sample-automated-reasoning-formalization", "external_url": "https://github.com/aws-samples/sample-automated-reasoning-formalization" }, { "id": "5023b165-dc45-4028-a9bc-5f817c55d336", "content_text": "> Architecture:\n- Multi-agent pipeline with orchestrator, workers, and reviewers operating in a DAG workflow\n- MCP tool servers: filesystem, git, bash, Lean REPL/LSP, mathlib\n- Supports multi-node execution via SLURM\n- Configurable LLM backends (Claude Opus 4.6, GPT, Gemini)", "date_published": "2026-05-29T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Autoform Bot: Multi-agent system for translating LaTeX mathematics into verified Lean 4 proofs", "url": "https://github.com/facebookresearch/autoform-bot", "external_url": "https://github.com/facebookresearch/autoform-bot" }, { "id": "b6ad50a4-e385-4ba8-893f-ba524f530907", "content_text": "> Key quotes from Levie (via X/LinkedIn, reported across multiple outlets):", "date_published": "2026-05-29T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Aaron Levie: AI Psychosis and the Last Mile of Agent Work", "url": "https://www.linkedin.com/posts/boxaaron_take-whatever-number-of-people-you-thought-share-7466141190086909953-aEGw/", "external_url": "https://www.linkedin.com/posts/boxaaron_take-whatever-number-of-people-you-thought-share-7466141190086909953-aEGw/" }, { "id": "0ce8d134-c983-4e83-9aed-15931345f8fb", "content_text": "> Scale (May 2026):\n- 26 books\n- 630,999 total lines of code (483,917 lines of Lean, excluding comments/blanks)\n- 46,203 declarations, 42,837 proved (92.7%)\n- 2,855 / 4,007 statements formalized (71.3%)\n- 183,157M tokens consumed", "date_published": "2026-05-29T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "ATLAS: Autoformalized Textbook Library At Scale", "url": "https://github.com/facebookresearch/atlas-lean", "external_url": "https://github.com/facebookresearch/atlas-lean" }, { "id": "e8aa4fa8-080d-4ba2-bb18-34b8b16c51f8", "content_text": "", "date_published": "2026-05-28T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Various LLM Smells", "url": "https://shvbsle.in/various-llm-smells/", "external_url": "https://shvbsle.in/various-llm-smells/" }, { "id": "ae18d303-975c-4f64-a4ef-ee1fd166e454", "content_text": "> (LinkedIn source blocked; synthesized from Columbia Business School event, Fortune reporting, and Forbes coverage of Chatterji's research)", "date_published": "2026-05-28T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Economic Paradoxes of AI - Ronnie Chatterji (OpenAI Chief Economist)", "url": "https://www.linkedin.com/pulse/economic-paradoxes-ai-aaron-ronnie-chatterji-hoixc/", "external_url": "https://www.linkedin.com/pulse/economic-paradoxes-ai-aaron-ronnie-chatterji-hoixc/" }, { "id": "9f5e2ee2-9a06-4bdc-8f75-51e279916e37", "content_text": "> Key features:\n- Live presence for agents and humans in the same document\n- Comments, suggestions, and provenance tracking\n- Integrates via MCP/skill install with Claude Code, Codex, OpenClaw\n- Designed for pre-implementation work: scoping, planning, spec review\n- Free, no login required\n- Uses `X-Agent-Id` headers for agent identity in presence", "date_published": "2026-05-28T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Proof Editor - Collaborative Document Editor for Agents and Humans", "url": "https://proofeditor.ai/?utm_source=everywebsite", "external_url": "https://proofeditor.ai/?utm_source=everywebsite" }, { "id": "cd960c1c-fc2c-4bf0-8bbf-4587948daf11", "content_text": "> Core idea: Recursive Reasoning Models (RRMs) use repeated computation to refine a persistent latent state rather than append new elements to an output sequence. This decouples reasoning depth from both parameter scale and output length: a compact model can perform many steps of internal computation by repeatedly applying shared transition functions", "date_published": "2026-05-28T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Generative Recursive Reasoning (GRAM)", "url": "https://arxiv.org/html/2605.19376v1", "external_url": "https://arxiv.org/html/2605.19376v1" }, { "id": "fc959e45-0084-4c9c-a861-2be9629d4a13", "content_text": "> The problem AI-assisted engineering amplifies:\n\"The prompt you give to the agent is the de-facto requirement now. Every vague prompt produces a vague spec or plan, and the AI agent implementing that spec produces code full of undisclosed decisions made on your behalf, without your awareness or agreement.\"", "date_published": "2026-05-27T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Requirements analysis: catching requirement bugs before they become code", "url": "https://kiro.dev/blog/deep-spec-analysis/", "external_url": "https://kiro.dev/blog/deep-spec-analysis/" }, { "id": "425768e1-2ae0-47cf-93ec-d4e94d22ca79", "content_text": "> Marking the 135th anniversary of Rerum novarum, Pope Leo XIV releases his first encyclical, entitled 'Magnifica humanitas: On Safeguarding the Human Person in the Time of Artificial Intelligence.' He appeals for the safeguarding of humanity, promotion of truth, dignity of work, social justice, and peace.", "date_published": "2026-05-25T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Magnifica Humanitas On Safeguarding the Human Person in the Time of Artificial Intelligence", "url": "http://www.vatican.va/content/leo-xiv/en/encyclicals/documents/20250515-magnifica-humanitas.html#The_res_novae_of_our_time", "external_url": "http://www.vatican.va/content/leo-xiv/en/encyclicals/documents/20250515-magnifica-humanitas.html#The_res_novae_of_our_time" }, { "id": "f2af21a5-096c-47e4-91ff-f132a037353f", "content_text": "> Stats: 34.7k stars, 2.8k forks, TypeScript 70.4%, MIT License", "date_published": "2026-05-24T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - Lum1104 Understand-Anything interactive knowledge graph for codebases", "url": "https://github.com/Lum1104/Understand-Anything", "external_url": "https://github.com/Lum1104/Understand-Anything" }, { "id": "6b5da41b-a0a2-4332-94d3-2b81d732a81b", "content_text": "> The LinkedIn post was not fully crawlable, but references the gist at https://gist.github.com/aartraju/cedca245d76894ebe44ba2322ab15682 which provides the full tutorial. See the companion expanded note for the gist content.", "date_published": "2026-05-24T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Aarthi Raju Amazon Quick productivity workflow LinkedIn", "url": "https://www.linkedin.com/posts/aarthi-raju_amazonquick-productivity-share-7463822542940573696-XFd-/", "external_url": "https://www.linkedin.com/posts/aarthi-raju_amazonquick-productivity-share-7463822542940573696-XFd-/" }, { "id": "ff65fb2f-8459-4b06-ba88-d732bc19d186", "content_text": "> AWS MCP Server\n- Full AWS API coverage (300+ services, 15,000+ API actions) through a single tool\n- Sandboxed script execution (isolated Python, no local filesystem/network access)\n- Real-time documentation access (search current AWS docs, guides, API references)\n- Enterprise controls: CloudWatch metrics, IAM context keys, CloudTrail audit logging", "date_published": "2026-05-23T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Agent Toolkit for AWS", "url": "https://aws.amazon.com/products/developer-tools/agent-toolkit-for-aws/", "external_url": "https://aws.amazon.com/products/developer-tools/agent-toolkit-for-aws/" }, { "id": "258c733e-06fc-4952-b76c-b1de4a2d1ace", "content_text": "> Title: The Last Harness You'll Ever Build \nAuthors: Haebin Seong, Li Yin, Haoran Zhang, Zhan Shi \nSubmitted: April 22, 2026 (v1), revised May 1, 2026 (v3) \nSubject: Artificial Intelligence (cs.AI)", "date_published": "2026-05-23T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Last Harness Youll Ever Build arXiv", "url": "https://arxiv.org/abs/2604.21003", "external_url": "https://arxiv.org/abs/2604.21003" }, { "id": "612dd647-5435-4578-b097-dcd889963ac0", "content_text": "> The LinkedIn post was not directly crawlable. Based on the URL slug and surrounding context:", "date_published": "2026-05-23T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The job apocalypse conversation is infuriating Conor Grennan LinkedIn", "url": "https://www.linkedin.com/posts/conorgrennan_the-job-apocalypse-conversation-is-infuriating-activity-7463980612467601408-9yIv", "external_url": "https://www.linkedin.com/posts/conorgrennan_the-job-apocalypse-conversation-is-infuriating-activity-7463980612467601408-9yIv" }, { "id": "c9199226-ad61-4b0c-b8b9-32b14de2fc8d", "content_text": "", "date_published": "2026-05-18T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "YOU HAD ONE JOB AI (to replace all the jobs)! | Chris Gaun", "url": "https://www.linkedin.com/posts/gaun_you-had-one-job-ai-to-replace-all-the-jobs-activity-7460316695950262274-B6OU?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/gaun_you-had-one-job-ai-to-replace-all-the-jobs-activity-7460316695950262274-B6OU?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "acd37436-2021-415a-9de8-f938fa286d92", "content_text": "", "date_published": "2026-05-18T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Last weekend was a tale of two commencement speeches. Both in Arizona, both to similar graduates. Eric Schmidt was booed. Harrison Ford was embraced. I listened to both. There's a lesson here about\u2026 | \ud83c\udf00 Patrick Copeland | 18 comments", "url": "https://www.linkedin.com/posts/patrickcopeland_last-weekend-was-a-tale-of-two-commencement-ugcPost-7462179765613928448-DWle?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/patrickcopeland_last-weekend-was-a-tale-of-two-commencement-ugcPost-7462179765613928448-DWle?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "b43cccfc-0f22-4cf3-bca1-20b571c2dd13", "content_text": "", "date_published": "2026-05-18T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The most interesting thing in tech: If you are going to give a commencement address in the next couple of weeks, do not mention AI or you might get loud boos. It happened in Florida last week; and it\u2026 | Nicholas Thompson | 70 comments", "url": "https://www.linkedin.com/posts/nicholasxthompson_the-most-interesting-thing-in-tech-if-you-ugcPost-7462259441602764801-TrCh?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/nicholasxthompson_the-most-interesting-thing-in-tech-if-you-ugcPost-7462259441602764801-TrCh?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "b54e0439-c8f6-4b61-9a8f-541d861e6fd7", "content_text": "", "date_published": "2026-05-17T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "#aiagents #skillos #selfevolvingai #skills #hermesagent | Melanie (Peiyao) Li", "url": "https://www.linkedin.com/posts/peiyaoli_aiagents-skillos-selfevolvingai-activity-7461869470693736448-H9VT?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/peiyaoli_aiagents-skillos-selfevolvingai-activity-7461869470693736448-H9VT?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "a84d2d3f-62ad-4b2a-ac49-b688ceedcb3c", "content_text": "", "date_published": "2026-05-14T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How AWS Is Using Neurosymbolic AI to Make Kiro More Reliable", "url": "https://theaieconomy.substack.com/p/aws-kiro-neurosymbolic-ai", "external_url": "https://theaieconomy.substack.com/p/aws-kiro-neurosymbolic-ai" }, { "id": "6ca0fb21-5a01-463c-b378-c3e60cf048e1", "content_text": "", "date_published": "2026-05-12T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Four security principles for agentic AI systems | AWS Security Blog", "url": "https://aws.amazon.com/blogs/security/four-security-principles-for-agentic-ai-systems/", "external_url": "https://aws.amazon.com/blogs/security/four-security-principles-for-agentic-ai-systems/" }, { "id": "6d785185-61b2-4422-aee2-fe5e65769410", "content_text": "", "date_published": "2026-05-12T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "There will be no AI jobpocalypse. The story that AI will lead to massive unemployment is stoking unnecessary fear. AI \u2014 like any other technology \u2014 affects jobs, but telling overblown stories of\u2026 | Andrew Ng | 443 comments", "url": "https://www.linkedin.com/posts/andrewyng_there-will-be-no-ai-jobpocalypse-the-story-share-7460002145052909568-qDiF?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/andrewyng_there-will-be-no-ai-jobpocalypse-the-story-share-7460002145052909568-qDiF?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "8b5396a2-0895-4f6f-9111-47234755a6f3", "content_text": "", "date_published": "2026-05-12T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - statewright/statewright: State machine guardrails for AI agents \u00b7 GitHub", "url": "https://github.com/statewright/statewright", "external_url": "https://github.com/statewright/statewright" }, { "id": "049cff83-731b-4ada-a746-684b1d175150", "content_text": "", "date_published": "2026-05-10T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Wave Terminal \u2014 Upgrade Your Command Line", "url": "https://www.waveterm.dev/", "external_url": "https://www.waveterm.dev/" }, { "id": "03123786-6a4d-49fe-88ce-11d634540ab7", "content_text": "", "date_published": "2026-05-10T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - Imbad0202/academic-research-skills: Academic Research Skills for Claude Code: research \u2192 write \u2192 review \u2192 revise \u2192 finalize \u00b7 GitHub", "url": "https://github.com/Imbad0202/academic-research-skills", "external_url": "https://github.com/Imbad0202/academic-research-skills" }, { "id": "2c043c9f-0d02-446b-8870-2bf99d93cf2a", "content_text": "", "date_published": "2026-05-10T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "This Anthropic engineer wrote the monumental blog post \"Building Effective Agents\". In just 14 minutes, he teaches you how to build AI agents the right way - more than most developers figure out in\u2026 | Linas Beli\u016bnas | 107 comments", "url": "https://www.linkedin.com/posts/linasbeliunas_this-anthropic-engineer-wrote-the-monumental-ugcPost-7459243468595421185-9guH?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/linasbeliunas_this-anthropic-engineer-wrote-the-monumental-ugcPost-7459243468595421185-9guH?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "04c30b21-bc76-4b91-a4c3-17302a312632", "content_text": "", "date_published": "2026-05-08T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - ageneralai/ageneral-agents-go \u00b7 GitHub", "url": "https://github.com/ageneralai/ageneral-agents-go", "external_url": "https://github.com/ageneralai/ageneral-agents-go" }, { "id": "34c669c3-c043-4e6f-acec-c4f299fcec57", "content_text": "", "date_published": "2026-05-07T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "#aws #agentcore #agenticai #mcp | Melanie (Peiyao) Li", "url": "https://www.linkedin.com/posts/peiyaoli_aws-agentcore-agenticai-activity-7458306582402080768-MmOd?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc", "external_url": "https://www.linkedin.com/posts/peiyaoli_aws-agentcore-agenticai-activity-7458306582402080768-MmOd?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc" }, { "id": "5d7432a8-1076-409f-9dcb-a8142a268204", "content_text": "", "date_published": "2026-05-06T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Secure AI Agents on the Desktop: What AWS Just Changed", "url": "https://www.cloud-bridge.co.uk/blog/secure-ai-agents-on-the-desktop-what-aws-just-changed", "external_url": "https://www.cloud-bridge.co.uk/blog/secure-ai-agents-on-the-desktop-what-aws-just-changed" }, { "id": "6a5976eb-3a0d-416a-9f7e-cdd13ec3a734", "content_text": "", "date_published": "2026-05-05T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Three Inverse Laws of AI - Susam Pal", "url": "https://susam.net/inverse-laws-of-robotics.html", "external_url": "https://susam.net/inverse-laws-of-robotics.html" }, { "id": "72314945-fdea-4e72-ac91-7c16346ad3a1", "content_text": "", "date_published": "2026-05-04T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Bullets of vaguely silvery hue", "url": "https://www.b-list.org/weblog/2023/oct/20/silver/", "external_url": "https://www.b-list.org/weblog/2023/oct/20/silver/" }, { "id": "096a071c-8652-4511-af9e-f817071a18b8", "content_text": "", "date_published": "2026-05-04T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Import AI 455: AI systems are about to start building themselves.", "url": "https://importai.substack.com/p/import-ai-455-automating-ai-research", "external_url": "https://importai.substack.com/p/import-ai-455-automating-ai-research" }, { "id": "9ae61a43-f40a-47f0-af16-5d119ac878f2", "content_text": "", "date_published": "2026-04-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "The Angine de Poitrine Argument for UBI", "url": "https://www.scottsantens.com/the-angine-de-poitrine-argument-for-ubi/", "external_url": "https://www.scottsantens.com/the-angine-de-poitrine-argument-for-ubi/" }, { "id": "18baadd9-19b2-4e77-9dd3-2ca26a0b7e62", "content_text": "", "date_published": "2026-04-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Harness engineering: leveraging Codex in an agent-first world | OpenAI", "url": "https://openai.com/index/harness-engineering/", "external_url": "https://openai.com/index/harness-engineering/" }, { "id": "f842ea41-7297-4711-9859-3ef555481619", "content_text": "", "date_published": "2026-04-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - marcbrooker/stability-sim: Experimental web-based simulator for exploring metastable behaviors in distributed systems \u00b7 GitHub", "url": "https://github.com/marcbrooker/stability-sim/", "external_url": "https://github.com/marcbrooker/stability-sim/" }, { "id": "65bc06e7-d50d-44d4-9532-b822eb212ab3", "content_text": "", "date_published": "2026-04-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - openai/symphony: Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents. \u00b7 GitHub", "url": "https://github.com/openai/symphony", "external_url": "https://github.com/openai/symphony" }, { "id": "b8ca06f4-ca83-4ae0-bc28-493b9f3bfe1d", "content_text": "", "date_published": "2026-04-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Introducing Strands Agents TypeScript 1.0: Build Production Agents in TypeScript | Strands Agents SDK", "url": "https://strandsagents.com/blog/strands-agents-typescript-v1/", "external_url": "https://strandsagents.com/blog/strands-agents-typescript-v1/" }, { "id": "2f616306-44eb-4f4f-99bc-2970e1245251", "content_text": "", "date_published": "2026-04-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "An open-source spec for Codex orchestration: Symphony. | OpenAI", "url": "https://openai.com/index/open-source-codex-orchestration-symphony/", "external_url": "https://openai.com/index/open-source-codex-orchestration-symphony/" }, { "id": "1fc2757c-ce73-46a0-9d1f-fc5be64b30d1", "content_text": "", "date_published": "2026-04-30T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "It's time to be right. - Marc's Blog", "url": "https://brooker.co.za/blog/2026/04/30/be-right.html", "external_url": "https://brooker.co.za/blog/2026/04/30/be-right.html" }, { "id": "9303f3ce-bc3c-4b2c-b2c0-5f42fafab1ff", "content_text": "", "date_published": "2026-04-29T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Home | Laws of UX", "url": "https://lawsofux.com/", "external_url": "https://lawsofux.com/" }, { "id": "ccb2bed4-81cf-4461-a70a-c7af4c17bf7a", "content_text": "", "date_published": "2026-04-27T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "From developer desks to the whole organization: Running Claude Cowork in Amazon Bedrock | Artificial Intelligence", "url": "https://aws.amazon.com/blogs/machine-learning/from-developer-desks-to-the-whole-organization-running-claude-cowork-in-amazon-bedrock/", "external_url": "https://aws.amazon.com/blogs/machine-learning/from-developer-desks-to-the-whole-organization-running-claude-cowork-in-amazon-bedrock/" }, { "id": "4dffdef1-3e9d-401d-b482-62b35ede48be", "content_text": "", "date_published": "2026-04-27T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Tolaria \u2014 A second brain for the AI era", "url": "https://tolaria.md/", "external_url": "https://tolaria.md/" }, { "id": "ed589e06-d126-4b5a-bfc9-c0c6bea50717", "content_text": "", "date_published": "2026-04-26T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - thedotmack/claude-mem: A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future sessions. \u00b7 GitHub", "url": "https://github.com/thedotmack/claude-mem", "external_url": "https://github.com/thedotmack/claude-mem" }, { "id": "e41feba9-c6b1-4429-a4c4-edfd0227f692", "content_text": "", "date_published": "2026-04-25T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "WebMCP: Making Every Website a Tool for AI Agents", "url": "https://www.arcade.dev/blog/web-mcp-alex-nahas-interview/", "external_url": "https://www.arcade.dev/blog/web-mcp-alex-nahas-interview/" }, { "id": "37035327-61ee-4298-85eb-c5a9619f6ff8", "content_text": "", "date_published": "2026-04-25T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Copernican Revolution - Wikipedia", "url": "https://en.wikipedia.org/wiki/Copernican_Revolution", "external_url": "https://en.wikipedia.org/wiki/Copernican_Revolution" }, { "id": "f5d7894f-f912-4c3f-a3bc-e9d712c6e059", "content_text": "", "date_published": "2026-04-25T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - webmachinelearning/webmcp: \ud83e\udd16 WebMCP \u00b7 GitHub", "url": "https://github.com/webmachinelearning/webmcp", "external_url": "https://github.com/webmachinelearning/webmcp" }, { "id": "d35941cf-d6e9-4428-9a43-a85ecc76e98d", "content_text": "", "date_published": "2026-04-24T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Get to your first working agent in minutes: Announcing new features in Amazon Bedrock AgentCore | Artificial Intelligence", "url": "https://aws.amazon.com/blogs/machine-learning/get-to-your-first-working-agent-in-minutes-announcing-new-features-in-amazon-bedrock-agentcore/", "external_url": "https://aws.amazon.com/blogs/machine-learning/get-to-your-first-working-agent-in-minutes-announcing-new-features-in-amazon-bedrock-agentcore/" }, { "id": "b25b3040-7149-46cc-a0cd-c0c2b750ee3a", "content_text": "", "date_published": "2026-04-24T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "DeepSeek_V4.pdf \u00b7 deepseek-ai/DeepSeek-V4-Pro at main", "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek_V4.pdf", "external_url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek_V4.pdf" }, { "id": "b3f0e61b-fc66-48da-842e-859506be2605", "content_text": "", "date_published": "2026-04-24T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "How can we develop transformative tools for thought?", "url": "https://numinous.productions/ttft/", "external_url": "https://numinous.productions/ttft/" }, { "id": "6db7b2a3-0ceb-4570-8b0e-5f2174a51d54", "content_text": "", "date_published": "2026-04-23T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Work with the garage door up", "url": "https://notes.andymatuschak.org/Work_with_the_garage_door_up", "external_url": "https://notes.andymatuschak.org/Work_with_the_garage_door_up" }, { "id": "8792dff7-c5e9-4468-bdc1-35b3e2533ef6", "content_text": "", "date_published": "2026-04-23T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Anti-marketing, after Michael Nielsen", "url": "https://notes.andymatuschak.org/zF9ywLHqHfN5rFuPApiyqmP", "external_url": "https://notes.andymatuschak.org/zF9ywLHqHfN5rFuPApiyqmP" }, { "id": "0c6b6b96-0891-4ab6-8f4a-10879d4c7044", "content_text": "", "date_published": "2026-04-22T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "GitHub - google/agents-cli: The CLI and skills that turn any coding assistant into an expert at creating, evaluating, and deploying AI agents on Google Cloud. \u00b7 GitHub", "url": "https://github.com/google/agents-cli", "external_url": "https://github.com/google/agents-cli" }, { "id": "fe16ccb9-1f8b-437d-859a-49f6e26628ef", "content_text": "", "date_published": "2026-04-21T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Kimi Vendor Verifier", "url": "https://www.kimi.com/blog/kimi-vendor-verifier", "external_url": "https://www.kimi.com/blog/kimi-vendor-verifier" }, { "id": "b205be91-0541-47fd-97e3-2e0de4aba88c", "content_text": "", "date_published": "2026-04-19T12:00:00+00:00", "_fyi": { "type": "link", "tags": [] }, "title": "Gas Town: from Clown Show to v1.0 | by Steve Yegge | Apr, 2026 | Medium", "url": "https://steve-yegge.medium.com/gas-town-from-clown-show-to-v1-0-c239d9a407ec", "external_url": "https://steve-yegge.medium.com/gas-town-from-clown-show-to-v1-0-c239d9a407ec" } ] }