What is mattwood.fyi?

A live, structured feed of riffs and links from Matt Wood, Chief AI & Technology Officer at AWS — readable by people and queryable by agents as a typed knowledge graph.

Is there an API and is it free?

Yes. A public, read-only query API is available with no authentication and no API keys. See https://mattwood.fyi/openapi.json and https://mattwood.fyi/docs.

How should an AI agent use this site?

Fetch https://mattwood.fyi/feed.json for the full structured feed, or query the API for keyword/semantic search and the connection graph. See https://mattwood.fyi/agents/ for when-to-use guidance.

Context Language Models

Language models that manage their own context by treating it as an editable file, enabling dynamic context updates and multi-agent systems while outperforming existing context management strategies across various tasks. The approach supports both in-context and parametric learning through natural language steering and reinforcement learning, with optimized serving via suffix cache reuse.

arxiv.org →

Connections

Related to

Skillsync - Portable AI Context for Coding Agents Both address portable/manageable AI context for coding agents — Skillsync provides portable context across agents while Context LMs treat context as an editable file for dynamic management, targeting overlapping problems
Infinite-Parameter LLMs: Generating Weights from Live Data Both explore non-static model architectures — infinite-parameter LLMs generate weights from live data while Context LMs treat context as live editable state, both pushing beyond fixed-parameter inference paradigms

Challenges

Unreal Agent: Cost-Efficient AI Agent Harness Context LMs' built-in dynamic context management challenges the need for external agent harness cost-efficiency solutions by internalizing context optimization rather than relying on harness-level strategies

Supports

Thinking Before Thinking: Scaling Agentic Inference Through Meta-Reasoning Context Language Models' dynamic context management complements meta-reasoning by enabling agents to adaptively update what information is in scope before and during inference, enhancing agentic reasoning pipelines
Mini-AGI: Continual Learning on 8GB VRAM Context Language Models' editable context and multi-agent support directly enables continual learning scenarios like Mini-AGI, where context must persist and evolve across learning episodes without full retraining

Type:	link
First published:	Thursday, 1 October 2026 at 17:57 UTC
Location:	https://arxiv.org/abs/2609.37725
URL:	https://mattwood.fyi/i/f7603ddc

I'm Matt Wood, and this is For Your Information. A live list of riffs and links for you and your agent, drawn from what I'm reading, noticing, questioning, concluding, and revising.

Connections