What it sells
Training grounds for AI agents.
Business model breakdown
Buyer: who owns the budget?
UnverifiedResearch and infrastructure leads at frontier and neo AI labs; AI/ML platform leaders at enterprises.
Trigger: what makes them buy now?
Agents are moving from demo to production and reliability is limited by the lack of realistic training and evaluation environments.
Workflow: what inputs become what output?
Build company-scale RL environments with real codebases and microservices to train agents; evaluate agents in environments mirroring customer systems; auto-optimize prompts and policies with GEPA and RL.
Economics: what it saves and how it charges
UnverifiedApplied research lab selling bespoke environments and optimization infrastructure to labs and enterprises (engagement-based); open work (OpenThoughts, Terminal-Bench, GEPA) builds credibility.
Distribution: how it reaches customers
Research-led, inbound 'Get in touch'; open-source datasets and benchmarks drive awareness.
Funding history and coverage
| Round date | Stage | Amount | Investors | Source |
|---|---|---|---|---|
| 2026-07-14 | — | $40M | — | news.google.com |
Profiles summarize public reporting and company websites. Business-model notes are MaxBC's reading of that material, not confirmed facts; unknowns are left blank rather than guessed. Updated 2026-10-12.