Half my time goes into using Claude Code, and the other half goes into optimizing my workflow for it.
What I have now:
• Everything organized by sprints in ./sprints/v1, v2 etc folder
• Custom command /prd to help me brainstrom requirements for a sprint and break it down into atomic tasks (which should take 5-10 mins each)
• Custom command /dev to pick highest priority task in a prd and follow test driven development to implement the tasks
• Custom command /walkthrough to write a sprint review report that details what code was produced so I can read and understand exactly what the code does
• I run Claude Code in docker with bypass permissions so it can run overnight without asking me anything. I use the excellent cco [1] for that!
• I made a Codex skill in claude so it can consult Codex for plans, tests, code review, etc, <- amazing what Codex catches that Claude doesn't
• Specify in /dev that testing has to happen via browser screenshots (so claude installs playwright and uses headless chromium for end to end testing)
Oh yes, btw, it helps that I have Claude Max $200/mo plan so basically I run everything on Opus and never worry about limits.
Now I feel I have a setup that automates 95% of whatever I want to build. I just brainstrom per sprint, and everything else happens automatically while I read research or live my life.
The only question now is – what to build? what to explore? everything seems possible.
[1] CCO https://github.com/nikvdp by @arghzero <- shoutout to him, he has added features very quickly to the repo. Super-responsive!
Join 200k followers
Follow @paraschopra