TypeSafe launched Jev because sequential LLMs are “totally useless for computers”

7.5 relevance

New startup challenging sequential LLM architecture for agents, highly novel and strategic.

Startup thenewstack.io

Answer the board on AI readiness

Arrive with evidence, not opinion. See how your foundation benchmarks against your peers, get a clear read, and leave with a ranked place to start.

Why It Matters

For engineers building multi-agent systems and decision workflows, Jev offers a deterministic, low-latency model that can reliably choose tools, verify outputs, and set guardrails—addressing the unpredictability and cost of sequential LLM calls.

Key Takeaways

Evaluate Jev for structured decision tasks in your agent orchestration stack to replace chat-based LLMs where speed, reliability, and cost are critical—especially for tool selection and confidence-based thresholds.

TypeSafe launched Jev because sequential LLMs are “totally useless for computers”

Summary

TypeSafe emerged from stealth with $40M in seed funding led by DCVC to launch Jev, a System One model that outputs typed decisions with calibrated probabilities instead of natural language—claiming 20-200x faster and 40-400x cheaper inference with no hallucinations. Built by ex-OpenAI co-founder Diogo Almeida, Jev uses a new training algorithm (RLCD) and is already seeing rapid adoption: within 24 hours on Vercel's AI Gateway, nearly 13% of paid teams used it, outpacing GPT and other models.

Author

Adrian Bridgwater