How to Get Your Org to Adopt Coding Agents (Without Shipping Garbage) — Eyal Blum, Figma
Source
youtube.com
Author
AI Engineer
Date
Context
Figma's approach to reluctant coding-AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → users starts with their knowledge of what can go wrong. Its conference talk argues that experienced engineers often carry undocumented knowledge about a codebase, making their objections useful evidence of missing checks. Giving those engineers responsibility for improving verification turns skepticism into concrete work on making agent output easier to trust.
The approach also addresses the attentionThe mechanism that lets a model weigh which earlier words matter for the word it's currently processing — the core operation of a transformer.Full definition → cost of generated text. The team describes opening each proposed code change with a human-written explanation, followed by generated detail. Reviewers can identify the author's own judgment before spending time on the longer analysis.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
attention — The mechanism that lets a model weigh which earlier words matter for the word it's currently processing — the core operation of a transformer.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
video1,000 Agent Tasks in a Sandbox: What Breaks When LLMs Run Code — Kevin Orellana, AWS
videoHow Inference Keeps Up with Our Insatiable Need for Tokens — Cerebras & Gimlet Labs
videoAuth for Agents: Let AI Sign Up for Your Product with auth.md — Michael Grinich, WorkOS
videoDecoupling the Brain from the Hands: Claude Managed Agents (Part 1 of 4) — Anthropic
