Microsoft paper reports higher scores and faster results with larger manager-free coding agent teams

Back Rohan Paul

Rohan Paul· @rohanpaul_ai · X· 2026-10-05 11:30 · Original publication time AI score54

Automatically verified and published · Generated and evidence-checked automatically; not reviewed by a human.

AI introduction

A Microsoft paper reports that, on the five hardest of ProgramBench’s 200 tasks, average scores rose at every step as a manager-free coding agent team scaled from 1 to 128 agents. In its Agensh approach, agents claim subtasks and coordinate through a shared Git repository and board; the paper does not report the cost of large teams.

Article · Original

The article text is unavailable in this language; an existing version is shown.

Turns out coding agents don't need a boss:

New Microsoft paper finds that bigger teams of coding agents score higher and get there sooner when agents claim their own tasks without a central manager.

Even on the 5 hardest of ProgramBench's 200 tasks, scaling a manager-free team from 1 to 128 agents raised the average score at every step.

so for big jobs, add agents and let them coordinate through shared tools with no lead agent.

Popular multi-agent coding tools send all work through 1 lead agent, which can only manage so many helpers.

Agensh drops the lead agent. Each agent claims a sub-task, builds and tests it, merges it into a shared Git repo, and logs findings on a shared board.

Every run used 1 model, and the paper does not report what large teams cost.

Found an error? Send a correction