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.

Six Months Writing Code Exclusively With AI Agents

Connections

Challenges

Why Junior Engineers Still Matter in the AI Era Six months of exclusively AI-written code suggests even experienced developers can delegate all coding to agents, potentially undermining the argument that junior engineers remain essential for foundational work

Supports

Agentic Coding: LLM Agents, Testing, and Hallucination Risks The six-month experience corroborates findings about agentic coding risks including hallucination and testing challenges, providing real-world longitudinal evidence for the theoretical concerns raised
AI Coding and the Bug Selection Problem Both address practical realities of AI-assisted coding from a developer perspective; the six-month experience directly illustrates the 'bug selection problem' where AI agents introduce subtle issues that require human judgment
Understanding is the new bottleneck The shift from writing code to directing agents reflects the thesis that 'understanding is the new bottleneck' — the developer's role becomes comprehending what to build rather than how to implement it
AI is removing the middle class of software engineering The developer's shift from writing code to directing AI agents directly illustrates the thesis that AI is restructuring software engineering roles, removing traditional middle-layer coding work

Related to

8 Myths of Software Development with AI The first-hand experience over six months serves as empirical evidence that either confirms or refutes common myths about AI software development, grounding the debate in longitudinal practice

Supported by

AI-Driven Development Life Cycle: Reimagining Software Engineering Both examine AI's transformation of software development workflows; the lifecycle framework provides theoretical context for the practitioner experience of coding exclusively with AI agents
Slideops: Repository Slide Deck Generator Six months of AI-only coding revealed how quickly code outpaces understanding; Slideops automates the conversion of repos into navigable decks to help developers maintain comprehension of AI-generated codebases
OpenCodex: Universal LLM Provider Proxy A proxy enabling any LLM with Codex CLI directly supports workflows where developers write code exclusively with AI agents, allowing them to choose preferred models without tooling constraints

Building a Software Factory: Merging 1000 PRs in a Week Both are first-hand practitioner accounts of AI agents handling real coding workloads, but the Software Factory scales beyond individual experience to team-level pipeline automation

Challenged by

How To Write With An LLM The new item's cautious, human-centered approach to LLM writing assistance contrasts with six months of exclusive AI-agent-driven coding, representing different philosophies on AI collaboration depth

Type:	link
First published:	Thursday, 27 August 2026 at 15:22 UTC
Location:	https://blog.exe.dev/engineering-with-ai
URL:	https://mattwood.fyi/i/079565ac

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