What it sells
Reinforcement Learning (RL) for AI Agents
Business model breakdown
Buyer: who owns the budget?
UnverifiedHeads of AI/ML at companies running agents or document extraction.
Workflow: what inputs become what output?
Customer data and tools go into a reinforcement fine-tuning pipeline (GRPO, DAPO, multi-turn tool training) with auto-retraining when eval scores drop; models served single-tenant or exported.
Economics: what it saves and how it charges
UnverifiedPlatform plus services; pitch is task-specific models that beat foundation models at a fraction of cost. No prices shown.
Distribution: how it reaches customers
Demo-led and forward-deployed: "Book a Demo" and hands-on engagement from dataset curation to delivery.
Funding history and coverage
| Round date | Stage | Amount | Investors | Source |
|---|---|---|---|---|
| — | Y Combinator Winter 2025 | — | Y Combinator | ycombinator.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.