40x More Merged PRs With 2.7x the Engineers — Walden Yan, CPO, Cognition
Cognition's merged PRs grew 40x in a year while its engineering team grew 2.7x.
Walden Yan explains what changed: self-driving PRs by default, small pods that own everything from bug reports to QA, and a workflow run almost entirely from Slack. He covers running every service locally so agents can test, VMs rebuilt every hour, and agents that optimize the factory itself. Then he shows the next step: giving agents jobs instead of tasks, from tuning inference on new GPUs to an Android app maintained by Devin over 44 days. Speaker info: - https://cognition.ai Timestamps: 0:00 - 40x merged PRs in a year 1:42 - Self-driving by default 2:31 - Small pods with vertical ownership 3:16 - Working entirely from Slack 4:46 - Every service runs locally for agents 6:20 - iPads and cloud-only security 7:05 - Agents that improve the factory 7:50 - Give agents jobs, not tasks 9:24 - Measure everything, let agents optimize 10:09 - Tuning inference on new GPUs 10:54 - An Android app in 44 days 13:13 - Be ambitious
Texten är källans egen beskrivning av publiceringen. Innehållet tillhör AI Engineer.
Mer att läsa
What Runs Where? Local AI in Production — Ahmad Osman, Osmantic
AI Engineer 21 nov.
Experts say AI has made planning cyberattacks on industrial equipment faster by helping attackers read manuals and research past accidents more quickly
Techmeme för 2 tim sedan
Current and former officials from the US and other countries describe the Chinese spy agency MSS' campaign to help Chinese companies acquire advanced AI chips
Techmeme för 6 tim sedan
Tesla rolls out FSD v15, featuring a photorealistic rendering of car surroundings; Elon Musk says it is real-time AI video generation without using game engines
Techmeme för 6 tim sedan
