Replicas

Run any coding agent harness in the cloud

2.8Score#408 of 999 agents#17 of 40 in Software engineer

Why it ranks #408

Replicas, an AI software engineer from YC Sp26, ranks #408 of 999 AI agents and #17 of 40 in Software engineer, with a score of 2.8 of 10. Data as of 3 Oct 2026.

The score adds four parts, each 0 to 10, by weight. Every point comes from public evidence. How scoring works ·

Proof 2 × 30% 0.6
Scale 1 × 30% 0.3
Momentum 3 × 25% 0.8
Autonomy 8 × 15% 1.2
= 2.8

Proof
Do customers use it? 30% of the score
2
Claims a customer count.
- +2
  A customer count
  “We’re used by 20+ YC startups at all team sizes.”
  Launch YC post · 1 Apr 2026
Not found: revenue or growth (+4), named customers (+3), a customer result (+2), work done at scale (+2)
Scale
Is there a real company behind it? 30% of the score
1
2 people.
- +1
  2 people on its YC page2 people
Not found: funding round in its news (+3)
Momentum
Is it shipping and growing now? 25% of the score
3
Shows hiring and a new batch.
- +2
  Hiring: 1 open roleFounding EngineerYC page
- +1
  New: YC Spring 2026
Not found: launch YC post in the last 6 months (+3), new traction claim in the last 6 months (+2), funding news in the last year (+2)
Autonomy
How much of the job does it do on its own? 15% of the score
8
Autonomous worker: completes whole tasks; people handle exceptions.
- +8
  Level 3 of 4: Autonomous workerCompletes whole tasks; people handle exceptions.
What the model read
- Acts in other systems: 90% likely
- Finishes whole tasks: 79% likely
- Calls itself an agent: 88% likely
Level 3: Completes whole tasks; people handle exceptions.

About Replicas

Replicas is a cloud coding agent that allows you to delegate tasks to Claude Code or Codex from the app, Slack, Linear, and more. Each task works inside of a sandboxed virtual machine, allowing teams to heavily parallelize work and burn through tickets.

Replicas agents output higher quality PRs via feedback loops - each task automatically is notified and fixes CI failures and reads all code reviews. With Environments, engineering teams can configure agents to run their entire codebase locally, verify its work, and provide preview URLs.