I started using terminal coding agents around June last year. I used Claude Code for a few months, then switched to Opencode because the experience was rough. My terminal would constantly flicker, formatting was messy, and a single session could drag my whole machine down. Then after a month or two I moved to Pi and have been using it ever since.
What’s strange is that after going through all these coding agents, my workflow hasn’t really changed. It’s still very primitive. I ask a question or show agent some ideas, discuss the approach, let the agent write code, then I review and verify. I did try to adopt some of the “high-efficiency” workflows people talk about, but none of them work for me.
Around the end of last year I saw Boris, the Claude Code Author, share how he uses Claude Code. I was shocked. He runs multiple sessions in parallel. I tried to replicate that. It didn’t work. Two parallel sessions already feel like my limit. Once I go beyond that, my brain starts falling apart within minutes. Context switching is painful, I will lose myself in ten minutes.
Recently I saw another tweet from him, talking about how to use newer features in Claude Code. I realized I hadn’t even heard of half of them. I honestly feel like I’m falling behind
I wonder you guys really running multiple parallel sessions and using all these newer features effectively, and on top of that double your productivity? Or is this kind of workflow only realistic for a small group of people, like those building the tools themselves?
The main reason is because if there's a significant bug or large optimization going on, that shit needs to be done, tested and merged before building more stuff on top, otherwise you run the risk of wasted time, tokens and effort having a bunch of parallel work running that may not end up compatible at the end.
Lately I've had a lot more success having Claude generate a plan, send the plan to Codex for co-validation/amendments, have Claude implement the plan, then have Codex PR review the commit (and likely make some edits of its own), then I test out the code/changes.
Meanwhile, my actual management of what I'm asking them to do is just a text file in Notepad where I'll write like BUG: xyz thing does abc or IDEA: let's change this to that as I'm testing in-app, with the actual code opened in Notepad++ tabs (lol feel free to roast me, I'm in front of 2 screens, one Windows (primary), one Mac (to the right), sharing keyboard and mouse -- LLMs are 99% on the Mac, planning/testing/verification/manual coding/graphic design on Windows, committing and pushing to a repo both machines have checked out)
I haven't yet found a scenario where many Claudes and many Codexes running simultaneously on 35 concurrent features makes any sense, but I'd definitely encourage people to try multi-model cooperation since they all seem to have different sensibilities. I haven't made much use of Gemini in this context though because two's company, three's a crowd. YMMV.
I expected this to become less necessary over time as models got faster, but the opposite has happened. It feels like Claude has actually gotten slower (but in fairness does more per prompt), meaning worktrees are even more essential now.
I think whats important is, that you keep atomical small tasks and increments, and whenever possible merge things. to many hanging worktrees can quickly also become a nightmare managing
It's not perfect; I've had some issues with Claude Code forgetting where it did things ("oh... it's not working because I'm not in the right directory"). I think it needs some architectural tweaks to function more reliably.