I’ve written software for about 30 years. I've been a heavy coding agent user for the past 1+ year. What practical coding-agent questions can I help answer?
Heat trend
Collecting trend data
The percentage is based on available heat signal, not comment count or independent people.
A software developer with 30 years of experience, including over 20 years of hands-on coding, has been a heavy user of AI coding agents for more than a year. They started with non-agentic tools like Aider, then moved to Claude Code, Droid, and Codex, often running them in tmux. More recently, they developed an in-house orchestration tool for Codex and Pi. They are offering to answer practical questions about coding agents or suggest solutions to problems.
I've been mostly hands on coding professionally for 20+ years. I have taken time in between to lead teams, run product management or run enterprise pre-sales.
I started with AI coding agents before they were agentic (Aider; it's still great) and, then Claude Code, Droid and finally Codex all running in tmux. I switch a few months ago to an inhouse orchestration tool I build for myself that runs Codex (app server) and most recently, Pi.
Feel free to ask me questions about coding with agents or not, running them in parallel, review, testing, error handling, etc etc.
Or if you have a problem that I might suggest a solution for, go for it too.