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.

Ornith-1.5: Self-Scaffolding to Self-Improvement

Connections

Supports

The Economics of Recursive Self-Improvement Ornith-1.5's self-scaffolding and continuous task generation is a concrete technical implementation of the recursive self-improvement economics discussed in 554B2536
The Shape of Things to Come, Part 1: The Continuous Thunderdome Ornith-1.5's continuous self-improvement loop via RL is a technical mechanism underlying the 'continuous thunderdome' competitive model improvement dynamic described in 9A5B38D2

Related to

Inkling: Thinking Machines' Open-Weights Model (and Tinker Fine-Tuning Platform) Both involve open-weights models with fine-tuning/improvement platforms; Ornith-1.5's self-improvement pipeline relates to Inkling's Tinker fine-tuning approach
Building an Advanced Agentic Harness Both address advanced agentic harness construction; Ornith-1.5 auto-generates scaffolds while 3FA0663B focuses on manually building advanced agentic harnesses
SIA: Self Improving AI Framework Both are self-improvement AI frameworks; FF88FEE8 is a self-improving AI framework while Ornith-1.5 specifically focuses on end-to-end self-scaffolding to generate tasks and rollouts for RL

Develops into

Managing Agent Skills with Dr. Skill Ornith-1.5's task-specific scaffold generation is a more automated evolution of the skill management concept explored in Dr. Skill for managing agent capabilities

Path to Astra Ornith-1.5's self-scaffolding to self-improvement concept mirrors the iterative capability development described in Astra's roadmap toward increasingly autonomous AI advancement
Microduck: Tiny Biped Robot Ornith-1.5's self-scaffolding to self-improvement parallels Microduck's capability to learn new behaviors - both systems are designed to expand their capabilities beyond initial programming
Security Cameras to Bird Identification with BirdNet-Go Both items operate in the ornithological/bird domain; 'Ornith-1.5' naming suggests bird-related AI, directly paralleling BirdNet-Go's automated bird classification mission
GDM Science Skills: Agentic Scientific Workflows Ornith-1.5's self-scaffolding approach to agent improvement shares architectural goals with GDM Science Skills' efficiency improvements, both seeking agents that progressively improve task execution
Harness Engineering Paper Collection Ornith-1.5 describes a self-scaffolding to self-improvement system, directly exemplifying the 'self-modifying systems' endpoint of the harness evolution timeline traced in the paper collection
WikiSkill: Agent Experience into Persistent Knowledge Both involve agents improving themselves through accumulated experience - Ornith-1.5 uses self-scaffolding for self-improvement while WikiSkill consolidates execution experience into reusable knowledge

Supported by

Harness Engineering Ornith-1.5's self-scaffolding to self-improvement directly requires the progressive hardening and architectural constraints described in harness engineering to prevent runaway self-modification
Teaching Qwen to Paint with Code Both apply self-improvement/RL-style training loops to enhance model capabilities beyond base supervised learning

Type:	link
First published:	Wednesday, 19 August 2026 at 19:49 UTC
Location:	https://ornith.ai/ornith_1_5.html
URL:	https://mattwood.fyi/i/286593b8

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