yibie/awesome-jev
A curated list of public projects, integrations, and discussions built on Jev — TypeSafe AI's System One model for typed decisions.
awesome-jev is a curated list of public projects and practices built on Jev, a model from TypeSafe AI that makes typed decisions — like a choice, a score, or a yes/no answer — each with a confidence level. Unlike a chat model, Jev takes unstructured input plus a specific question and returns a structured decision, making it useful for tasks like classification, routing, scoring, and guardrails in software.
The list answers two questions: where is Jev actually being used in real workflows, and which decision patterns carry across industries. It is organized into categories such as Classification & Routing, Verification & Guardrails, Agent Decisions, Infra/SDKs, Finance, and more. Each entry lives in exactly one category, chosen by its direct application domain. There is also a category for related discussions (X threads, Reddit, interviews) when no strong standalone project page exists.
To be included, an entry must be public and citable, genuinely use Jev (or a documented port) for a concrete decision task, explicitly name Jev or show a typed-decision loop, and have a one-sentence summary. Generic classifiers, pure theory, launch hype, and private sources are excluded.
Important caveat: Inclusion is not an endorsement. The list only checks that an entry meets the inclusion rules — it does not review code quality, security, maturity, or whether the project even runs. The README warns especially about same-day bulk submissions that share a scaffold and thin commit history; these can satisfy every rule yet be unproven. It provides a checklist for adopters: verify the code actually calls the Jev API, look for a runnable test or demo, trace any numbers to a source, check how much of the repo is code vs. prose, and confirm a license exists.
The README itself is generated from separate category files, so it serves as the homepage aggregate. It also lists projects tagged by coding agent (e.g., Claude Code, Pi, Codex) and includes a full list of entries per category, though the excerpt only shows a few examples from Classification & Routing.
Related
- Project
- Project
- Project
- Project