Winged Victory of Samothrace · Louvre · CC0the wing · y = 0.00061x² − 1.82x + 1667 · head not recoveredFor AI labs
RL environments for agents that handle money.
Built to order: an agent is given a budget and has to pay for what the task needs. The reward comes from the settlement ledger, not a judge: who it paid, how much, how often, whether it stayed within budget, and whether its own report matches the ledger. We synthesize every task and golden trajectory and write every grader; no customer data goes in.
What we build
Built to order · all synthesized · no customer data
| Task family | Graded by |
|---|---|
| Buy what the task needs, under a budget | Ledger: who was paid, how much, budget kept |
| Retry a failed call without paying twice | Ledger: duplicate payments |
| Choose the right seller from a catalog | Ledger: payee · code: deliverable fields |
Examples of scope, agent payments first. Each environment is built for one task family and comes with a checkable outcome and a cost per task run.
How an environment is made
- 01 · Tasks · synthesized
- 02 · Environments · agents that pay
- 03 · Graders · the ledger, not a model
- 04 · Delivered · licensed to you
You get · The environment, the task set, the graders, reviewed golden trajectories, a calibration report and a statement of rights.