lean-collab
AI
multi-agent orchestration
SHOW HN2026
How HN took it
4 positive, 10 neutral or asking, 0 mixed, 5 critical, out of 19 comments about the project. Thread score -0.04.
Idea: 4 for, 2 againstExecution: 0 for, 1 againstDesign: 0 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Clean approach of bounded blast radius and code-encoded policy
From 1 comment
Great work! I like the approach of maximum freedom inside bounded blast radius and how you use code to encode policy.
visarga on HN
Lean verification makes the feedback loop objective
From 1 comment
The Lean angle here is really interesting: most multi-agent demos dodge hard verification, but tying each agent’s output to makes the feedback loop objective. Curious how you’re…
raphaelmolly8 on HN
Objections
One well-tuned agent is enough for most tasks; multi-agent adds overhead
From 2 comments
Even Anthropic research articles consistently demonstrate they themselves use one agent, and just tune the harness around it. I ignore all Skills, MCPs, and view all of these as…
gck1 on HNThats the other thing, you hit the nail on the head, I dont want 20 agents unless they're doing research and scouring code. Claude can do that just fine. I want Claude Code doing…
giancarlostoro on HN
Hard to observe many agents at once
From 1 comment
Even Anthropic research articles consistently demonstrate they themselves use one agent, and just tune the harness around it. I ignore all Skills, MCPs, and view all of these as…
gck1 on HN
Requires a proprietary Ensue API key
From 1 comment
seems like it requires an API key to your proprietary Ensue memory system
slopusila on HN
Requests
Add clearer signup and API key recovery flow
From 1 comment
When you get a chance to work on your login flow, I recommend giving users an opportunity to request the key rather than automatically showing it once only on the first screen. I…
yodon on HN
Suggest workloads to test next
From 2 comments
Math proofs are really easy to run with this specific harness. Our next experiments are going to be bigger, think full code base refactors. We're working on applying RLM to…
austinbaggio on HNCool, what’s a good first task to try this on where it’s likely to beat a single agent?
christinetyip on HN
Offline open-model setup for VPS
From 1 comment
Impressive! Are there any boilerplates that people know of for running something similar to this using open offline models? Would be cool to run this (or a single agent version)…
hsdev on HN
Questions
How does progress subscription work
From 1 comment
“How does progress subscription work — are agents watching specific signals (test failures, TODO list, build status), or just a global feed?”
zmanian on HN
How are goal claim conflicts handled
From 1 comment
The Lean angle here is really interesting: most multi-agent demos dodge hard verification, but tying each agent’s output to makes the feedback loop objective. Curious how you’re…
raphaelmolly8 on HN
Is shared memory RAM or a markdown file
From 1 comment
So is this shared memory as in RAM or a markdown file that they update with their statuses?
arbol on HN
Alternatives named
Beads
From 2 comments
I feel like there's two camps: * Throw more agents * Use something like Beads I'm in the latter, I don't have infinite resources, I'd rather stick to one agent and optimize what…
giancarlostoro on HNThats the other thing, you hit the nail on the head, I dont want 20 agents unless they're doing research and scouring code. Claude can do that just fine. I want Claude Code doing…
giancarlostoro on HN
GasTown
From 1 comment
For the moment, researching multi-agent orchestration. At first glance, your work looks among the best in class of published work I've seen. Particularly interested to understand…
yodon on HN
Claude Code swarms
From 1 comment
For the moment, researching multi-agent orchestration. At first glance, your work looks among the best in class of published work I've seen. Particularly interested to understand…
yodon on HN
Ralph loop
From 1 comment
It's a more of a black box with claude, at least with this you see the proof strategy and mistakes made by the model when it decomposes the problem. I think instead of Ralph…
miligauss on HN