mattwood.fyi

How Organizations Use ChatGPT

2026-08-13T23:51:30Z

https://cdn.openai.com/pdf/how-organizations-use-chatgpt.pdf

This is all seems pretty straightforward, but this caught my attention:

active use within adopting firms spans job functions and seniority levels, with especially high usage intensity among early-career workers

This wasn't the case a few years ago, when high-tenure workers were dominant.

Understanding is the new bottleneck

2026-08-13T23:49:09Z

https://www.geoffreylitt.com/2026/07/02/understanding-is-the-new-bottleneck

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.

Bullet: Fast Coding Agent

2026-08-13T19:41:10Z

https://www.codewithbullet.com/

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.

Lots happening in speed, latency, and efficiency.

Kimi K3: Complete Developer Guide for 2026

2026-08-13T19:38:16Z

https://www.firecrawl.dev/blog/kimi-k3

A comprehensive guide to Kimi K3, covering its features and capabilities for developers in 2026.

GPT-5.6 Sol Ultrafast: Frontier Intelligence at Unprecedented Speed

2026-08-13T19:36:43Z

https://www.cerebras.ai/blog/accelerating-gpt-5-6-sol-ultrafast-with-openai

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.

Introducing Gemini 3.7 Flash

2026-08-13T19:36:19Z

https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/

Google announces Gemini 3.7 Flash, a high-performance AI model designed to deliver intelligent results with efficient processing capabilities for practical applications.

Ordinary Abundance

2026-08-13T17:23:52Z

https://ordinaryabundance.com/

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.

Fun visual story telling (and a history I hadn't really thought about this way before)

Comparing 11 Different AI Models

2026-08-13T15:31:35Z

https://www.netlify.com/blog/one-prompt-11-models-very-different-results/

Exploration of 11 AI models including DeepSeek, Qwen, Kimi and open-source alternatives, examining their different outputs and capabilities for the same prompts.

Why AI Agents' Deceptive Behavior Concerns Users

2026-08-13T15:30:40Z

https://www.economist.com/business/2026/08/12/ai-agents-lie-cheat-and-steal-that-is-putting-off-users

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.

DeepSeek Harness: Plugin Architecture

2026-08-13T15:29:42Z

https://github.com/deepseek-ai/deepseek-harness

A plugin-based system where components are built as plugins, enabling extensible architecture for DeepSeek AI applications.

Amazon's AI Shopping Assistant Transforms Retail with Conversational Search

2026-08-13T15:22:22Z

https://x.com/amazonnews/status/2087637165500960905

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.

Material Discovery Bench: LLM Research Benchmark

2026-08-12T23:36:25Z

https://discoveredmaterials.com/research/

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.

Grok 4.6 Benchmarks and Cost Efficiency Analysis

2026-08-12T22:55:47Z

https://artificialanalysis.ai/articles/grok-4-6-benchmarks-and-analysis

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.

Cedar Policy

2026-08-12T20:44:15Z

https://cedarpolicy.com/en

Cedar is an open-source policy language and evaluation engine designed for authorization and access control decisions in applications and services.

Now is an excellent time to learn more about Cedar.

Introducing Dogwood: runtime verification for AI agents

2026-08-12T20:43:24Z

https://aws.amazon.com/blogs/opensource/introducing-dogwood-runtime-verification-for-ai-agents/

Dogwood is a runtime verification framework designed to monitor and ensure the safe and correct behavior of AI agents during execution.

The Hardest Working Font in Manhattan

2026-08-12T19:37:48Z

https://aresluna.org/the-hardest-working-font-in-manhattan/

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.

Exquisite storytelling for font nerds (and everyone else).

Book Prize Index

2026-08-12T19:33:56Z

https://book-prize-index.vercel.app/

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.

Fun way to find 'what to read next'.

Qwen 3.8 2.4T Mixture of Experts Model

2026-08-12T19:31:58Z

https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B

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.

DeepSeek V4 Pro 0813 API Pricing & Benchmarks

2026-08-12T19:31:27Z

https://openrouter.ai/deepseek/deepseek-v4-pro-0813

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.

AI is removing the middle class of software engineering

2026-08-12T15:49:01Z

https://blog.florianherrengt.com/ai-removing-middle-class-software-engineering.html

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.

What Sort of Maths Are LLMs Good At?

2026-08-12T15:48:32Z

https://gowers.wordpress.com/2026/08/12/what-sort-of-maths-are-llms-good-at/

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.

OpenAI Models Now Available on Amazon Bedrock

2026-08-11T23:29:25Z

https://www.aboutamazon.com/news/aws/bedrock-openai-models

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.

ChatGPT Desktop App for Linux Now in Preview

2026-08-11T23:02:02Z

https://x.com/OpenAI/status/2087231350134980830?s=20

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.

Emergent Introspective Awareness in Large Language Models

2026-08-11T23:01:25Z

https://arxiv.org/abs/2601.01828

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.

The Economics of Recursive Self-Improvement

2026-08-11T23:01:05Z

https://elasticity.institute/rsi-paper.pdf

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.

Nemotron 3.5 Lightning and NeMo Switchyard for Agentic AI

2026-08-11T22:57:34Z

https://blogs.nvidia.com/blog/nemotron-lightning-switchyard-rtx-dgx/

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.

Compression is prediction

2026-08-11T22:57:18Z

https://ngrok.com/blog/compression-is-prediction

Explores the fundamental relationship between data compression and predictive modeling, examining how compression algorithms work as predictors and the implications for AI and LLMs.

How Keras 3 Modernized Expedia's Lodging Ranking Stack

2026-08-11T19:56:00Z

https://medium.com/expedia-group-tech/how-keras-3-helped-modernise-expedia-groups-lodging-ranking-stack-7fec96f052fd

Expedia Group upgraded their lodging ranking system using Keras 3, improving their machine learning infrastructure for hotel search results.

The Creative Power of Invisibility

2026-08-11T19:46:50Z

https://www.linkedin.com/pulse/creative-power-invisibility-oana-leonte-j1erc/

Explores how restraint, curation, and subtraction—exemplified by producer Rick Rubin's invisible hand in shaping iconic music—represent the true creative power in an era of algorithmic noise and constant visibility.

Write.md - Customizable Markdown Editor for macOS

2026-08-11T19:45:05Z

https://writemd.app/

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.

I love my growing collection of Markdown editors. This one looks fun.

MiniMax H3 Inference Engine for Mac

2026-08-11T19:43:38Z

https://github.com/antirez/h3.c

An H3 inference engine implementation for Mac computers, providing MiniMax-based AI inference capabilities.

OpenSSH 10.5 Release Notes

2026-08-11T18:43:51Z

https://www.openssh.org/releasenotes.html#10.5

Recently the OpenSSH team have received a large number of security bug reports, many of which are findings from AI models or made with AI assistance. While many AI reports are determined not to have security impact when considered in the context of a realistic threat model, we very much welcome these reports, especially when combined with human triage, analysis, test-cases and particularly when accompanied by proposed fixes.

We have seen a number of cases where a security bug identified by AI tools is subsequently independently discovered by a different researcher. This suggests that adversaries who do not report bugs to OSS projects are likely to be able to discover these bugs too. Given this, the OpenSSH team will, for now, be making more frequent releases to get bugfixes into users' hands more quickly rather than batching them until the next planned release.

Makes sense.

11–16× Faster LLM Inference with llama.cpp

2026-08-11T17:55:25Z

https://github.com/trycua/cua/blob/main/blog/gpu-passthrough-macos-vms.md

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.

LiquidAI/LFM2.5-2.6B · Hugging Face

2026-08-11T17:13:54Z

https://huggingface.co/LiquidAI/LFM2.5-2.6B

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.

Cactus Needle 2

2026-08-11T01:10:01Z

https://cactuscompute.com/needle

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.

Feels 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.

What's the best programming language for coding agents?

2026-08-11T01:09:52Z

https://danluu.com/pl-tokens/

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.

Parametron

2026-08-11T01:09:41Z

https://ethw.org/Milestones:Parametron,_1954

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.

Japan’s first university-built stored-program computer which became the nation's then-fastest in 1958

50,000 boat names

2026-08-11T01:04:30Z

https://www.beautifulpublicdata.com/boat-names/

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.

An important dataset, but the all-time best boat name of all time is clearly: Earn Trussed.

How This Was Made

2026-08-11T00:16:21Z

https://mattwood.blog/essays/2026/08/how-this-was-made/

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 — not just sharing outputs — is key to organizational change.

Humanising LLM Outputs is Dumb

2026-08-10T23:04:28Z

https://kuber.studio/blog/Reflections/Humanising-LLM-Outputs-is-Actually-Dumb

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.

Herdr is joining Y Combinator. The runtime stays open.

2026-08-06T21:23:01Z

https://herdr.dev/blog/herdr-is-joining-y-combinator/

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.

Taste Is All That's Left

2026-08-06T21:20:10Z

https://notashelf.dev/posts/taste-is-all-thats-left

An essay arguing that as AI collapses the cost and effort of building software, the filtering function that effort once provided disappears, leaving "taste" — the judgment of what deserves to exist — as the remaining meaningful differentiator in software craft.

AMD acquires AI chip startup Taalas to boost inference performance by etching models into silicon

2026-08-06T21:06:45Z

https://www.theregister.com/systems/2026/08/06/amd-acquires-ai-chip-startup-taalas-to-boost-inference-performance-by-etching-models-into-silicon/5284344

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.

Improving Gpt 5 6 Sol In Chatgpt

2026-08-06T19:46:04Z

https://openai.com/index/improving-gpt-5-6-sol-in-chatgpt/

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.

Discovery Loop

2026-08-05T17:13:16Z

https://www.discoveryloop.com/

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.

Building an Advanced Agentic Harness

2026-08-05T15:41:30Z

https://data4sci.com/blog/building-an-advanced-agentic-harness

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.

LLMs can't jump

2026-08-05T15:40:06Z

https://openreview.net/forum?id=klU4737opt

TL;DR: Scientific invention requires manipulative abduction and physical simulation

Intelligence is not the main bottleneck

2026-08-05T15:38:14Z

https://www.writingruxandrabio.com/p/intelligence-is-not-the-main-bottleneck

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.

8 Myths of Software Development with AI

2026-08-05T01:04:10Z

https://queue.acm.org/detail.cfm?id=3807963

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.

The myths are:

Developers Spend Most of Their Time Writing Code
Writing Code Is the Bottleneck
Lines of Code Written by AI Is the Best Measure of Impact
AI Helps All Tasks and Engineers Equally
AI Will Turn Individual Developers into 10x Developers
It’s Up to Each Developer to Make AI Work
High-Performing AI Tools Will Be Adopted Automatically
With GenAI, Enterprises Can Innovate at Startup Speed

Pi, Minimal and Performant

2026-08-05T01:02:47Z

https://earendil.com/posts/pi-autoresearch-and-databricks/

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.

Pi 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.

Most tech revolutions made work worse for employees. AI could be the exception: this+that

2026-08-04T23:00:00Z

https://www.thisandthat.chat/blog/most-tech-revolutions-made-work-worse-for-employees/

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.

Why some people mow a lawn better than others

2026-08-04T22:13:09Z

https://pudding.cool/2026/06/mow/

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.

From the article, a good description of algorithms and heuristics.

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 “good enough” path, fast.

Also, this domain is a keeper.

Kiro Crew

2026-08-04T18:15:41Z

https://kiro.dev/crew/

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 — 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.

The Shape of Things to Come, Part 1: The Continuous Thunderdome

2026-08-04T18:11:57Z

https://yegge.ai/essays/the-shape-of-things-to-come/

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.

Don't be special, stay out in front, and you will see the future clear as day.

Good advice.

TerminalWidget for Mac, iPhone, and iPad

2026-08-04T18:03:47Z

https://terminalwidget.app/

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.

Great way to enable your agent to share status, notifications, data, etc. Cool.

Marin

2026-08-04T01:32:51Z

https://marin.community/

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.

Open Athena

2026-08-04T01:31:52Z

https://openathena.ai/

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.

Leiden Declaration on Artificial Intelligence and Mathematics

2026-08-03T20:52:56Z

https://leidendeclaration.ai/

qm

2026-07-31T23:54:39Z

https://github.com/yc-software/qm

A multiplayer agent harness for work. In Slack and on the web.

How AgentCore Gateway Supports the MCP 2026-07-28 Spec

2026-07-28T12:00:00Z

https://aws.amazon.com/blogs/machine-learning/how-agentcore-gateway-supports-the-mcp-2026-07-28-spec/

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.

Does AI eliminate jobs? Economists find heavy adopters hire more.

2026-07-23T12:00:00Z

https://ramp.com/data/ai-jobs-impact

Companies that invest heavily in AI grow headcount 10% over the two years following adoption. Entry-level headcount grows 12%.

Fireworks AI: Specialized Intelligence Infrastructure

2026-07-20T12:00:00Z

https://fireworks.ai/

Jensen Huang positioning: "Fireworks is the TSMC of AI Factories..."

Strands AI Functions: Python Library for Verified Agent Workflows

2026-07-20T12:00:00Z

https://github.com/strands-labs/ai-functions/

America, 1926: What a Forgotten 100-Year-Old Report Says About Who We Are

2026-07-20T12:00:00Z

https://www.derekthompson.org/p/america-1926-an-absurdly-deep-dive

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).

Lovable: AI App Builder

2026-07-20T12:00:00Z

https://lovable.dev/

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.

Inertia-1: Unified Motion Foundation Model from Wearable Sensors

2026-07-20T12:00:00Z

https://yang-ai-lab.github.io/Inertia-1/

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."

Strands AI Functions: Post-Conditions and Multi-Agent Teams as Python Functions

2026-07-20T12:00:00Z

https://github.com/strands-labs/ai-functions

Post-conditions pattern:

@ai_function(post_conditions=[check_length, check_style], max_attempts=5)
def summarize_meeting(transcripts: str) -> MeetingSummary:
    """Write a summary of the following meeting in less than 50 words."""

Post-conditions can be plain Python assertions or other AI Functions. The function only returns once eve

Kiro CLI Custom Agents

2026-07-20T12:00:00Z

https://kiro.dev/docs/cli/custom-agents/

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.

The Uni-Context: A Philosopher's One-Word Theory to Explain Why the World Feels So Weird

2026-07-20T12:00:00Z

https://www.derekthompson.org/p/a-philosophers-one-word-theory-to

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.

Designing APIs for Agents - Freestyle

2026-07-20T12:00:00Z

https://www.freestyle.sh/blog/opinion/designing-apis-for-agents

Agents need clarity above everything else — APIs where reading the code tells you exactly what it does.

The Zero-Cost Fallacy: Open Source in the Agentic Era (Thoughtworks)

2026-07-20T12:00:00Z

https://www.thoughtworks.com/insights/blog/open-source/zero-cost-fallacy-open-source-agentic-era

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.

Databricks AI: Agent Bricks and Unity AI Gateway

2026-07-20T12:00:00Z

https://www.databricks.com/product/artificial-intelligence

Key positioning: "Stop relying on generic AI models. Databricks has the tools to build agent systems that deliver accurate, data-driven results."

Agent Swarms and the New Model Economics

2026-07-20T12:00:00Z

https://cursor.com/blog/agent-swarm-model-economics

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.

Valkey: Open Source High-Performance Key/Value Datastore

2026-07-20T12:00:00Z

https://valkey.io/

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.

Schema: Frontier Models with the Right Harness Achieve ~99% on ARC-AGI-3 Public

2026-07-20T12:00:00Z

https://schema-harness.github.io/

ARC-AGI-3 gives an agent a game environment, without an explanation of what it is seeing. At each step, the agent receives a 64×64 grid of 16 color indices and a set of legal actions. The environment supplies no object list, rule sheet, stated goal, or shaped reward.

Together AI: Research-Driven Inference and Training Platform

2026-07-20T12:00:00Z

https://www.together.ai/

Key research highlights at ICML 2026: - FlashAttention-4: Algorithm and kernel pipelining co-design for asymmetric hardware scaling (Tri Dao et al.) - Mamba-3: Next-generation state space model (CMU + Princeton + Together + Cartesia) - DeepSWE: Fully open-sourced state-of-the-art coding agent trained by scaling RL - Cache-aware prefill-decode disag

The Anti-Mac Interface

2026-07-15T12:00:00Z

https://www.nngroup.com/articles/anti-mac-interface/

Analyzing Metastable Failures

2026-07-15T12:00:00Z

https://www.amazon.science/publications/analyzing-metastable-failures

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

Inkling: Thinking Machines' Open-Weights Model (and Tinker Fine-Tuning Platform)

2026-07-15T12:00:00Z

https://thinkingmachines.ai/news/introducing-inkling/

Model specs: - 975B total params, 41B active (Mixture-of-Experts) - Inkling-Small: 12B active params - 1M token context window - Pretrained on 45T tokens (text, images, audio, video) - Controllable thinking effort (0.2 to 0.99 sweep)

Ways to think about token pricing

2026-07-09T12:00:00Z

https://www.ben-evans.com/benedictevans/2026/7/9/ways-to-think-about-token-pricing

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.

Bridging Intent and Execution in Agentic Systems

2026-07-08T12:00:00Z

https://www.amazon.science/blog/bridging-intent-and-execution-in-agentic-systems

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.

The New Network is You

2026-07-07T12:00:00Z

https://www.linkedin.com/pulse/new-network-you-merritt-robinson--6l68c?utm_source=share&utm_medium=member_android&utm_campaign=share_via

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.

Neuronpedia Jacobian Lens: Interactive J-Space Explorer for Qwen3.6-27B

2026-07-06T12:00:00Z

https://www.neuronpedia.org/qwen3.6-27b/jlens

Amazon at ICML 2026: Research Presence and Booth Schedule

2026-07-05T12:00:00Z

https://www.linkedin.com/pulse/amazon-icml-2026-papers-imen-grida-ben-yahia-ph-d--znhse

Sponsorship: Diamond level, Booth B207, Seoul South Korea

Micro-Agent: Beat Frontier Models with Collaboration inside Model API

2026-06-29T12:00:00Z

https://vllm.ai/blog/2026-06-29-micro-agent-frontier-models

Core thesis: "The phrase 'frontier model' is starting to mean two things. One is a checkpoint. The other is a system boundary."

AWS Bedrock AgentCore Skill - Claude Code Plugin

2026-06-27T12:00:00Z

https://github.com/ferdinandobons/AWSBedrockAgentCoreSkill

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.

Agent Session-Aware Runtime

2026-06-27T12:00:00Z

https://thenewstack.io/agent-session-aware-runtime/

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.

Generative artificial intelligence creates delicious, sustainable, and nutritious burgers

2026-06-26T12:00:00Z

https://www.nature.com/articles/s41538-026-00953-x

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.

gogcli spec: Unified Go CLI for Google Workspace

2026-06-26T12:00:00Z

https://gogcli.sh/spec.html

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

Sir Clive Sinclair: Tireless Inventor Ahead of His Time

2026-06-14T12:00:00Z

https://www.bbc.com/news/science-environment-29985976

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.

Formal Methods at Jane Street: Agentic Coding Changes the Calculus

2026-06-14T12:00:00Z

https://blog.janestreet.com/formal-methods-at-jane-street-index/?from_theconsensus=1

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.

MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling

2026-06-12T12:00:00Z

https://arxiv.org/abs/2606.13473

General-purpose large language models outperform specialized clinical AI tools on medical benchmarks

2026-06-12T12:00:00Z

https://www.nature.com/articles/s41591-026-04431-5

SIA: Self Improving AI Framework

2026-06-11T12:00:00Z

https://github.com/hexo-ai/sia

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

How Frontier Teams Are Reinventing AI-Native Development

2026-06-10T12:00:00Z

https://aws.amazon.com/blogs/machine-learning/how-frontier-teams-are-reinventing-ai-native-development/

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.

Brad Porter: We Saw This With Robotics at Amazon Too

2026-06-09T12:00:00Z

https://www.linkedin.com/posts/brad-porter-cobot_we-saw-this-with-robotics-at-amazon-too-share-7470168784201273346-XpL8/

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:

The Dynamo and the Computer: An Historical Perspective on the Modern Productivity Paradox

2026-06-09T12:00:00Z

https://www.almendron.com/tribuna/wp-content/uploads/2018/03/the-dynamo-and-the-computer-an-historical-perspective-on-the-modern-productivity-paradox.pdf

Key argument (from David, 1990, American Economic Review, Vol. 80, No. 2, pp. 355-361):

Loop Engineering - by Addy Osmani - Elevate

2026-06-09T12:00:00Z

https://addyo.substack.com/p/loop-engineering

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.

Lessons I've Learned at Warp Building for Developers

2026-06-04T12:00:00Z

https://www.linkedin.com/pulse/lessons-ive-learned-warp-building-developers-zach-lloyd-cbgoc

"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." — Zach Lloyd

Exploring AWS DevOps Agent Part 3 - Troubleshooting Issues

2026-06-03T12:00:00Z

https://awstip.com/exploring-aws-devops-agent-part-3-1d16b7f2dd72

Issue 1: Pod stuck in Pending state - Agent correctly identified missing Fargate profile for the test namespace.

How One Tech Company Created 13 New Types of Jobs Because of A.I.

2026-06-02T12:00:00Z

https://www.nytimes.com/2026/06/01/technology/box-13-new-types-jobs-ai.html

Key details from coverage (NYT article paywalled; sourced from aggregated reporting):

First They Built a Secular Apocalypse Belief System. Now They Want Religious Authority.

2026-06-02T12:00:00Z

https://www.aipanic.news/p/first-they-built-a-secular-apocalypse

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.

What's Easy Now? What's Hard Now? - Marc's Blog

2026-06-02T12:00:00Z

https://brooker.co.za/blog/2026/05/18/whats-easy-whats-hard.html

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.

Training Azerbaijani Language Models on Amazon SageMaker AI

2026-06-01T12:00:00Z

https://www.linkedin.com/posts/sabirmardan_amazon-sagemaker-amazon-share-7467195020136144897-lSzD/?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc

Problem: Adapting foundation models to a morphologically rich language (Azerbaijani) with limited training data, no existing blueprint for efficient LLM training in that language.

The Economic Implications of Learning by Doing - Kenneth Arrow (1962)

2026-06-01T12:00:00Z

https://www.haverford.edu/sites/default/files/Arrow1962.pdf

"Learning can only take place through the attempt to solve a problem and therefore only takes place during activity."

A Rational Conversation on Where AI Is Actually Going | Benedict Evans on Lenny's Podcast

2026-06-01T12:00:00Z

https://www.linkedin.com/posts/lennyrachitsky_my-biggest-takeaways-from-benedict-evans-share-7467220148777656320-CvQS/?utm_source=share&utm_medium=member_android&rcm=ACoAAACiYBcBLrSH5NDbPZzkgyYFUlhyh0ZuGwc

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

COLLEAGUE.SKILL: Automated AI Skill Generation via Expert Knowledge Distillation

2026-06-01T12:00:00Z

https://arxiv.org/abs/2605.31264

Authors: Tianyi Zhou, Dongrui Liu, Leitao Yuan, Jing Shao, Xia Hu

Ethan Mollick: Meaningfully Better AI Releases Are Accelerating

2026-05-31T12:00:00Z

https://www.linkedin.com/posts/emollick_it-does-seem-like-meaningfully-better-ai-share-7466647737384742912-WpIl/

Mollick's post (May 30, 2026): "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."

Automating Low-Risk Code Review at Meta: RADAR, Risk Calibration, and Review Efficiency

2026-05-30T12:00:00Z

https://arxiv.org/abs/2605.30208

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.

Kiro CLI 2.5: Thinking Display, Subagent Review Loops, and Display Controls

2026-05-30T12:00:00Z

https://kiro.dev/changelog/cli/2-5/

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.

ARchitect: Automated Reasoning Policy Formalization for Bedrock Guardrails

2026-05-29T12:00:00Z

https://github.com/aws-samples/sample-automated-reasoning-formalization

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.

Autoform Bot: Multi-agent system for translating LaTeX mathematics into verified Lean 4 proofs

2026-05-29T12:00:00Z

https://github.com/facebookresearch/autoform-bot

Architecture: - Multi-agent pipeline with orchestrator, workers, and reviewers operating in a DAG workflow - MCP tool servers: filesystem, git, bash, Lean REPL/LSP, mathlib - Supports multi-node execution via SLURM - Configurable LLM backends (Claude Opus 4.6, GPT, Gemini)

Aaron Levie: AI Psychosis and the Last Mile of Agent Work

2026-05-29T12:00:00Z

https://www.linkedin.com/posts/boxaaron_take-whatever-number-of-people-you-thought-share-7466141190086909953-aEGw/

Key quotes from Levie (via X/LinkedIn, reported across multiple outlets):

ATLAS: Autoformalized Textbook Library At Scale

2026-05-29T12:00:00Z

https://github.com/facebookresearch/atlas-lean

Scale (May 2026): - 26 books - 630,999 total lines of code (483,917 lines of Lean, excluding comments/blanks) - 46,203 declarations, 42,837 proved (92.7%) - 2,855 / 4,007 statements formalized (71.3%) - 183,157M tokens consumed

Various LLM Smells

2026-05-28T12:00:00Z

https://shvbsle.in/various-llm-smells/

Economic Paradoxes of AI - Ronnie Chatterji (OpenAI Chief Economist)

2026-05-28T12:00:00Z

https://www.linkedin.com/pulse/economic-paradoxes-ai-aaron-ronnie-chatterji-hoixc/

(LinkedIn source blocked; synthesized from Columbia Business School event, Fortune reporting, and Forbes coverage of Chatterji's research)

Proof Editor - Collaborative Document Editor for Agents and Humans

2026-05-28T12:00:00Z

https://proofeditor.ai/?utm_source=everywebsite

Key features: - Live presence for agents and humans in the same document - Comments, suggestions, and provenance tracking - Integrates via MCP/skill install with Claude Code, Codex, OpenClaw - Designed for pre-implementation work: scoping, planning, spec review - Free, no login required - Uses X-Agent-Id headers for agent identity in presence

Generative Recursive Reasoning (GRAM)

2026-05-28T12:00:00Z

https://arxiv.org/html/2605.19376v1

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

Requirements analysis: catching requirement bugs before they become code

2026-05-27T12:00:00Z

https://kiro.dev/blog/deep-spec-analysis/

The problem AI-assisted engineering amplifies: "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."

Magnifica Humanitas On Safeguarding the Human Person in the Time of Artificial Intelligence

2026-05-25T12:00:00Z

http://www.vatican.va/content/leo-xiv/en/encyclicals/documents/20250515-magnifica-humanitas.html#The_res_novae_of_our_time

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.

GitHub - Lum1104 Understand-Anything interactive knowledge graph for codebases

2026-05-24T12:00:00Z

https://github.com/Lum1104/Understand-Anything

Stats: 34.7k stars, 2.8k forks, TypeScript 70.4%, MIT License

Aarthi Raju Amazon Quick productivity workflow LinkedIn

2026-05-24T12:00:00Z

https://www.linkedin.com/posts/aarthi-raju_amazonquick-productivity-share-7463822542940573696-XFd-/

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.

Agent Toolkit for AWS

2026-05-23T12:00:00Z

https://aws.amazon.com/products/developer-tools/agent-toolkit-for-aws/

AWS MCP Server - Full AWS API coverage (300+ services, 15,000+ API actions) through a single tool - Sandboxed script execution (isolated Python, no local filesystem/network access) - Real-time documentation access (search current AWS docs, guides, API references) - Enterprise controls: CloudWatch metrics, IAM context keys, CloudTrail audit logging

The Last Harness Youll Ever Build arXiv

2026-05-23T12:00:00Z

https://arxiv.org/abs/2604.21003

Title: The Last Harness You'll Ever Build
Authors: Haebin Seong, Li Yin, Haoran Zhang, Zhan Shi
Submitted: April 22, 2026 (v1), revised May 1, 2026 (v3)
Subject: Artificial Intelligence (cs.AI)