Will Jev Replace LLM's? What is Jev From TypeSafe AI
Krish Naik explains TypeSafe AI's Jev decision model, contrasting its fast probabilistic classification with generative LLM reasoning loops. He walks through conceptual architectural diagrams showing how Jev can handle tool routing, multi-agent dispatch, and post-generation guardrail verification without replacing generative LLMs.
Original by Krish NaikAgent workflowsIntermediate23 min 27 secPublished Source reviewed
Before you press play
What you’ll find in the video
Krish presents Jev as a decision model that returns class probabilities alongside, rather than in place of, a generative model.
His routing diagram places a bounded classification before tool or model selection; the video does not measure the resulting savings.
A second diagram puts Jev after generation as an accept, reject, or regenerate gate, with task-specific validation still needed.
Worth knowing
Gemini-assisted video/transcript review. The presenter relies on TypeSafe marketing materials and high-level conceptual walk-throughs; reported speedups, cost figures, and low hallucination claims are unverified vendor numbers without live benchmark code.