The company’s pitch starts from a question Almeida puts at the top of the launch post: “Models have been superhuman at chat for years, so where is all the automation?” Its answer is a model that returns a typed decision your code can branch on, with a probability attached, instead of a paragraph you have to parse.
TypeSafe’s own team page says Almeida co-invented RLHF and InstructGPT and previously worked at Google Brain, Gafni is a repeat founder with production AI experience, and Sheng was a research engineer at Meta’s FAIR. That background matters mainly because the training method behind Jev, RLCD, is described as a sibling of RLHF aimed at calibrated probabilities rather than preferred text.
The product surface is small. One model, one endpoint, and three question types. Pricing is $0.042 per million input tokens with output free. Access is a closed managed API in early access, opened off a waitlist at launch, and the weights are not public.
Every performance figure on the site, including the claimed 70ms to 500ms response time, comes from TypeSafe’s own evaluations. Specs are collected on the Jev model page.