About
AI lab targeting developers who need decisions inside software, not text for humans to read. Its flagship model, Jev, skips text generation entirely: send a schema of typed questions plus unstructured state, get back constrained answers with calibrated probabilities in a single parallel pass. Trained with RLCD (Reinforcement Learning for Calibrated Decisions), Jev runs at 70, 500 ms per call.
Activity
2026
1 story so far in 2026
| Month | Stories |
|---|---|
| January | 0 |
| February | 0 |
| March | 0 |
| April | 0 |
| May | 0 |
| June | 0 |
| July | 0 |
| August | 0 |
| September | 0 |
| October | 1 |
| November | 0 |
Focus areas
1 story