Anil-matcha/awesome-jev-by-typesafe on GitHub
- Stars
- 830
- Forks
- 164
- Language
- Python
- License
- MIT
- Last push
- 23 Sep 2026
Read from GitHub on . Counts change daily.
This is a community-maintained list, not an official TypeSafe resource, and the README says so directly. It organizes use cases, patterns, prompts and starter code around what Jev is meant for: classify, route, score, detect, rank, extract, verify and gate automation, while keeping control flow, thresholds and side effects in the caller’s code rather than in the model.
The repository predates Jev: its commit history starts in May 2023.
It links out to sibling collections by the same author and to two linked projects with their own numbers: a technical SEO audit study that crawled 1,204 pages and ran 4,816 typed judgments per run, at $0.0048 for a 12-query triage batch, and a daily arXiv and bioRxiv paper radar that filtered 501 papers for $0.0196 in one measured run. Those figures belong to the linked projects, not to this list.
How it uses Jev
with TypeSafeClient() as client:
response = client.system_one(
state=state,
questions={
"intent": Choice(
instructions="What is the customer's main request?",
criteria={
"refund": "The customer wants money returned.",
"technical_help": "The customer needs a bug or integration fixed.",
"information": "The customer is asking for information only.",
"other": "None of the other options clearly fits.",
},
),
"is_urgent": Noul(
instructions="Does the ticket explicitly communicate time pressure?",
