Notes on the recent session we had related to autonomous long-running coding agents.
(bookmark it)
Topics: /goal, loop engineering, verifiers, dynamic workflows, and much more.
So much to unpack, so I tried to quickly summarize the most relevant parts using my writer agent.
http://x.com/i/article/2065876120965111808
Users respond positively to the session on autonomous long-running coding agents and dynamic workflows by offering supportive clarification and help with the rough notes.
Find the recording here:
More coming next week.
How to effectively run autonomous long-running coding agents?
This is one of the most exciting discussions on agents I've ever had.
I recorded it and am making it freely available.
(bookmark it)
The idea of autonomous long-running agents is a real thing.
We talk about lots of things like /goal, /loop, and dynamic workflows, and what comes next.
One interesting discussion was around how to make the agent run for longer while ensuring it stays on track.
Most models today will struggle to coordinate work effectively. They sometimes pause the work early. Lots of mistakes happen, and lots of weird shortcuts (reward hacking).
What helps is to be extremely clear about the goals it needs to achieve. To clarify the dos and don'ts clearly. Eliminate any assumptions you think the model would make. Deep expertise matters so much in this.
But you can get far through careful planning. My formula currently is to use Opus 4.8 for planning carefully and GPT-5.5 for all executions. For the evaluator (via /goal), I am often using something like Deepseek or the latest models from Qwen, Kimi, and MiniMax, etc.
Another insight we discussed to enforce goals is to provide strong visual cues for the agent to compare with. I found that a multimodal goal is a much stronger goal than a plain text one. And use agents to help you set clear goals.
Watch here: https://academy.dair.ai/events/cmplo7v3b000e04l1pxprat4d
If you see anything that you would like clarification on, please don't be afraid to ask away. These are just rough notes I put together to more clearly explain some of the ideas discussed during the session. I will take the time to write more about this stuff in time, along with actual things to try.
Find the recording here:
More coming next week.
@omarsar0 loop engineering and dynamic workflows are the parts i need to pay more attention to
feels like thats where most projects get stuck
@omarsar0 posting the summary is a good signal
how close is this to the real agent you'd let run autonomously?