Well, not sure why you need 3 developers, unless you can split work on really independent parts.
My pipeline:
- beads for task tracking.
- one Claude Code session that runs few development lanes. Each lane is dynamic workflow that
- takes a ready task
- runs a designer agent if there are ui changes
- runs implementation agent (on a git worktree)
- runs review agent
- loop to implementation once if it rejected
- if approved creates a PR
- when CI checks are green (tests, vulnurability audits etc) marks as lane-verified)
- task can be marked as needed my attention (when it needs some decision, or manual configuration/action)
- same session runs a lander, which
- take lane-verified task
- rebases on master
- ensures master is green
- merge
- deploy
- hand tasks back to pipeline if conflict resolution is complex or CI is red after rebase.
- I don't look at the code
- I don't verify individual tasks
- in the other session I review what needs my attention - that's actual bottleneck, sometimes questions are very stupid and can be saved by LLM directly.
- I have dedicated sessions for planning/researching big feature/epics.
- I review periodically how product is working, and submit bugs via my own tool (https://session-replay.com)
This delivers 50-100 PRs a day (depending on my ability to resolve questions) with a quite decent quality.
Apart of my attention, CI is other bottleneck (had to install a dedicated server for CI).
Serious questions:
- roughly how much are you spending on tokens to achieve all this?
- what's the longevity of the application/s you're developing? (i.e. are you maintaining the same codebase for 1 year+, or shipping one-off small applications frequently?)
My main concern with leaning in this hard is the long-term impact to complexity and maintainability, which coding agents have been shown to struggle with moreso than solving the problem now.
I use Claude Code Max and weekly limit drains in 2 days. In tokens that would be thousands.
Well, you cannot predict how development will look in 6-12 months. So I would not worry about it. Agenting coding gave 5x-10x boost, autonomous loops give another 5x-10x. Who knows what will be next.
"Don't worry about maintainability; by the time you get there it will be solved"? That's gambling the future of the company on the rate of improvement of AI. If the company is not in the AI space, that may not be a bet they want to make.
Well, not sure why you need 3 developers, unless you can split work on really independent parts.
My pipeline: - beads for task tracking. - one Claude Code session that runs few development lanes. Each lane is dynamic workflow that - takes a ready task - runs a designer agent if there are ui changes - runs implementation agent (on a git worktree) - runs review agent - loop to implementation once if it rejected - if approved creates a PR - when CI checks are green (tests, vulnurability audits etc) marks as lane-verified) - task can be marked as needed my attention (when it needs some decision, or manual configuration/action) - same session runs a lander, which - take lane-verified task - rebases on master - ensures master is green - merge - deploy - hand tasks back to pipeline if conflict resolution is complex or CI is red after rebase. - I don't look at the code - I don't verify individual tasks
- in the other session I review what needs my attention - that's actual bottleneck, sometimes questions are very stupid and can be saved by LLM directly. - I have dedicated sessions for planning/researching big feature/epics. - I review periodically how product is working, and submit bugs via my own tool (https://session-replay.com)
This delivers 50-100 PRs a day (depending on my ability to resolve questions) with a quite decent quality. Apart of my attention, CI is other bottleneck (had to install a dedicated server for CI).
Feel free to ask if you have questions.
Serious questions: - roughly how much are you spending on tokens to achieve all this? - what's the longevity of the application/s you're developing? (i.e. are you maintaining the same codebase for 1 year+, or shipping one-off small applications frequently?)
My main concern with leaning in this hard is the long-term impact to complexity and maintainability, which coding agents have been shown to struggle with moreso than solving the problem now.
I use Claude Code Max and weekly limit drains in 2 days. In tokens that would be thousands.
Well, you cannot predict how development will look in 6-12 months. So I would not worry about it. Agenting coding gave 5x-10x boost, autonomous loops give another 5x-10x. Who knows what will be next.
"Don't worry about maintainability; by the time you get there it will be solved"? That's gambling the future of the company on the rate of improvement of AI. If the company is not in the AI space, that may not be a bet they want to make.
Ah, one thing I missed: I started with a good boilerplate (happens also to be mine), so most architectural questions are already solved.
[flagged]