Real-SWE: A coding benchmark built from private company codebases
Can coding agents can do real engineering work on codebases they've never seen?
Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.
Want to reach these readers? Sponsor an issue.
YC · launched Sep 10, 2026 · 12 upvotes (top 38%) · 0 views
Niche: ai benchmarking platforms · 145 launches, 4 in the last 7 days ▼ -56% · Explore this niche →
Category and tags
- Horizontal
- Observability & eval
- Developer
- AI-native
- Commercial product
- Generative AI
- Reinforcement Learning
- B2B
- AI
Visit website ↗ Source listing ↗
More like this: all 145 launches in ai benchmarking platforms →
Could you build this?
Partial A web interface for displaying benchmark results is trivial, but sourcing, sanitizing, and sandboxing realistic engineering environments from proprietary commercial codebases requires specialized access and tooling.
What it would actually take: Building Real-SWE requires automated pipeline orchestration using Docker/Kubernetes to reliably reproduce complex legacy microservice architectures, build chains, and test suites across private enterprise repositories. Sourcing requires bilateral legal agreements with enterprise software companies and rigorous data scrubbing/anonymization workflows to protect IP. The execution engine must provide strict multi-tenant isolation and deterministic evaluation metrics for agentic code edits across diverse languages.
Competitors
Other products that read as similar to this one : 3951 launches clear the similarity bar, closest 8 shown.
Attention rank: #1464 of 3952 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).
Launched 316 days after the earliest competitor.
- sloppy · github · 2026-09-10 · 13 upvotes · similarity 0.73
- codeden · github · 2026-11-07 · 11 upvotes · similarity 0.72
- Zenbu - The extensible IDE for managing coding agents · yc · 2026-06-03 · 5 upvotes · similarity 0.71
- OpenSpec: The Spec Framework for Coding Agents · yc · 2026-03-06 · 78 upvotes · similarity 0.71
- Loom, an open-source delivery harness for coding agents · hn · 2026-06-10 · 5 upvotes · similarity 0.71
- PACE-Bench · github · 2026-10-20 · 22 upvotes · similarity 0.70
- agent-scaffold · github · 2026-10-20 · 11 upvotes · similarity 0.70
- Jcode: High-performance coding agents · yc · 2026-08-10 · 35 upvotes · similarity 0.70
The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.