I made a thing that shows Claude Code, Cursor and Codex every screen their change touched, so they fix what they broke before you see it.
Introducing UI Verify! https://t.co/v9WFkr7B7N
@franmoretti_@valasellis@jsmonthlylondon Be trying to keep it on the smaller side, the smaller the better still. Small meaning methods in a class in a single pr, theb API endpoints in a pr, then the UI. Oh and things like edge cases and race conditions and stuff are also in their own PRs
AI code reviews: start small, turn mistakes into rules, and build trust with proof. Igor Luchenkov shares a playbook you can use on Monday, with tools you already have.
A huge thank you to @ag_grid for sponsoring and hosting our event. See you there!
AI code reviews: start small, turn mistakes into rules, and build trust with proof. Igor Luchenkov shares a playbook you can use on Monday, with tools you already have.
A huge thank you to @ag_grid for sponsoring and hosting our event. See you there!
Introducing Querio
The first analytics platform that puts your entire data stack in a git repo you own
Querio makes your agents 100% consistent on your data
Comment 'data teams' and i'll send you our open-source repo to move your BI tool to a git repo you own
@rohanbhanot@AnatoliKopadze I donβt think they look at the code anymore at that scale. I donβt do it for my project. What I look at instead is a high level description of the change and if it sounds good I let it go. Thats the only way to do it IMO with an army of agents
My Claude Code now goes "hmm" like a Minecraft villager every time it needs me.
One agent: charming.
Three agents: concerning.
Six agents: I live in a village now and they all want my emeralds.
Sound on for the video π
@Moha__niang Playwright MCP is cool but need to also do the visual testing to make sure nothing is broken across the whole app. Wrote about it here: https://t.co/qiXbdjctwi