---
title: "Discussion on \"What it took for five coding agents to beat one\" | Hashnode"
description: "Discussion on \"What it took for five coding agents to beat one\". One coding task, three setups, 18 runs, and five findings about what made the split pay off"
url: "https://hashnode.com/posts/what-it-took-for-five-coding-agents-to-beat-one/6b0270e6599bb579eea06592"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [What it took for five coding agents to beat one](https://blog.albedoweb.com/what-it-took-for-five-coding-agents-to-beat-one?utm_source=hashnode&utm_medium=feed)

One coding task, three setups, 18 runs, and five findings about what made the split pay off

14h ago · 14 min read · blog.albedoweb.com  
Read the article

-   [#ai-agents](https://hashnode.com/tag/ai-agents)
-   [#llm](https://hashnode.com/tag/llm)
-   [#software-engineering](https://hashnode.com/tag/software-engineering)
-   [#ai](https://hashnode.com/tag/ai)

## Written by

[Nikolay Fokin](https://hashnode.com/@albedolab)

@albedolab

  
[View profile](https://hashnode.com/@albedolab)

Databases, AI agents, and things I build

## Responses 4

Markdown supported · ⌘↵ to post

[Ahmet Özel](https://hashnode.com/@ahmetozel)

[9h ago](https://hashnode.com/posts/what-it-took-for-five-coding-agents-to-beat-one/6b0270e6599bb579eea06592/comment/6b02b913c8790508fabb4381)

Running six trials per setup against 30 hidden tests makes the 36-minute versus 24-minute comparison actually credible instead of a one-off anecdote. Giving five agents disjoint file ownership in separate repository copies avoids merge-conflict thrashing while keeping the integration step explicit.

What stands out most is that a single instruction change cut redundant work by more than 3x. In multi-agent coding, telling each worker the exact local check it owns—and what NOT to re-verify across the whole repo—often saves more wall-clock time than picking a faster model.

[Nikolay Fokin](https://hashnode.com/@albedolab)

[6h ago](https://hashnode.com/posts/what-it-took-for-five-coding-agents-to-beat-one/6b0270e6599bb579eea06592/comment/6b02b913c8790508fabb4381#reply-6b02e789eb096d2b575f848b)

Thanks! Agreed on the last point: in these runs the check line mattered more than the number of agents.

[AI Developer](https://hashnode.com/@lycore)

[13h ago](https://hashnode.com/posts/what-it-took-for-five-coding-agents-to-beat-one/6b0270e6599bb579eea06592/comment/6b028656695c13a5b580a419)

The finding that the same split won or lost depending on whether each agent got exact verify commands is a useful isolating result. With six runs per setup, though, the 18 to 61 minute auto-split spread makes me curious whether the win is the frozen plan or the narrower checks, since your own plan also carried the exact commands. A fourth arm with a frozen plan but the original loose VERIFY line, or the auto-split given your narrow-check template, would separate those two effects.

[Nikolay Fokin](https://hashnode.com/@albedolab)

[12h ago](https://hashnode.com/posts/what-it-took-for-five-coding-agents-to-beat-one/6b0270e6599bb579eea06592/comment/6b028656695c13a5b580a419#reply-6b028977397850ac97133aac)

Fair point, the two effects are mixed. The frozen plan with the loose VERIFY line exists as one trial run (62 min of agent work vs 26 with narrow commands), which is why it's the first item in my limits. Running the auto-split with my narrow-check template is a good next arm, I'll add it.
