TypeSafe AI exits stealth with $40M seed and System One model Jev

TypeSafe AI said it raised about $40 million in seed funding led by DCVC and opened early access to Jev, its first “System One” model built for fast, structured, type-safe decisions that software can call directly — with calibrated confidence scores — instead of open-ended chat text. Co-founders include Diogo Almeida (former OpenAI researcher, RLHF/ChatGPT co-inventor), Erik Gafni, and Sasha Sheng.

FINANCE desk — same-day primary seed for a frontier lab explicitly building non-chat models for production software automation, with a named OpenAI-alumni founder and DCVC lead.

Founders, as the Business Wire has it: Diogo Almeida, co-founder and CEO — a former OpenAI researcher the company calls a co-inventor of RLHF and ChatGPT methods — with Erik Gafni and Sasha Sheng. The same about-box says TypeSafe was founded in 2024 and is headquartered in San Francisco. RLHF means reinforcement learning from human feedback — the method that trained language models to follow instructions and chat. File those names, the CEO title, the 2024 founding, and the San Francisco HQ as TypeSafe’s. Extra team-page titles stay in Sources. This desk did not audit the bios.

First public model, still company: Jev, named for Jevons Paradox — the idea that making something cheaper can increase how much of it people use. TypeSafe calls Jev a “System One” model, a Kahneman-inspired label for fast, structured judgment rather than a chat assistant. The company says Jev is built for type-safe structured outputs software can consume directly, with calibrated confidence and probability scores on every answer, and that it gives up free-form string generation. Type-safe here means outputs must match a predefined schema so software will not crash on a weird string. File that Jev / System One / typed-decisions picture as TypeSafe’s. This desk did not call the model.

Speed and price, company claims — not independent Bad Signal benchmarks: the company blog says end-to-end latency is roughly 70 to 500 milliseconds, input pricing is about $0.042 per million tokens, and output tokens are free. A millisecond is a thousandth of a second. The same blog claims large speed and cost gains versus frontier chat models on System One-shaped tasks. File those figures as TypeSafe’s. This desk did not time a query or audit a bill.

Training and sampling, still company: TypeSafe names the training method Reinforcement Learning for Calibrated Decisions, or RLCD. Calibrated here means a higher confidence score is supposed to mean the model is actually right more often. The company says a parallel sampler generates all outputs in one query rather than token-by-token, the usual chat-model method of writing one word at a time. File RLCD and that parallel-sampler picture as TypeSafe’s. How the sampler is wired stays in Sources. This desk did not train a model.

Early access, company: TypeSafe opened a waitlist at typesafe.ai and says it is bringing developers off that list as it can. The company pitches “machine-native, composable AI” as a software primitive — a building block other programs can call — for classification, routing, scoring, extraction, and guardrails. Guardrails here means checks that score or block unsafe prompts and outputs. File the waitlist and that use-case list as TypeSafe’s. This desk did not join the waitlist.

Named voices on the company pages: Diogo Almeida, co-founder and CEO; and James Hardiman, general partner at DCVC. File the names and titles as theirs, via TypeSafe. Their quotes are color only and stay in Sources. This is not investment advice.

Plain English for the rest of the card: seed = early venture money. System One here = fast structured judgment, the company’s Kahneman-inspired label, not a chat assistant. type-safe = outputs must match a predefined schema so software will not crash on a weird string. RLCD = Reinforcement Learning for Calibrated Decisions. calibrated confidence = a probability that is supposed to match how often the model is actually right. Jevons Paradox = the idea that making something cheaper can increase how much of it people use.

PRIMARY here: TypeSafe’s 15 Sep 2026 company blog — Tier A PRIMARY company source, the original record for the System One / Jev product — plus the same-day Business Wire company syndication via MarketMinute, the funding record. Dual Tier A PRIMARY. The company site is product-home / waitlist context, not a second originating newsroom. The approximately $40 million seed, the DCVC lead, the Almeida / Gafni / Sheng founding, the 2024 San Francisco about-box, the Jev / System One / typed-decisions / calibrated-confidence picture, the 70–500 ms / $0.042-per-million-input / free-output claims, RLCD and the parallel sampler, the early-access waitlist and classification / routing / scoring / extraction / guardrails pitch, and the Hardiman name are company-attributed. Speed, cost, and “can’t hallucinate” lines stay company-attributed — not independently measured here. NOT claimed: a later Series letter, a post-money valuation, ARR, named enterprise customers, independently verified latency or price, that this desk tested Jev or saw a term sheet, a stock tip, or investment advice. Distinct from the already-filed mgi-visiomics-europe, scaffold-15m-seed, factory-200m-5b-valuation, zero-103m-seed, liquid-compute-15m-seed, profound-180m-series-d, isometric-ice-50m-series-a, and gensyn-open-1b-auditable.