Jev Parallel Dispatch
Assign simulated response teams from parallel Jev judgments.
What Jev decides
Which incident each team should address, plus risk and reassessment judgments.
Jev Parallel Dispatch asks one Choice per team, then uses a deterministic optimizer to respect incident capacity and apply assignments.
- Jev role
- Evaluate incident assignments and global pressure for simulated teams.
- Input
- The shared simulation snapshot, team capabilities, incidents, and operator doctrine.
- Output
- Per-team Choice distributions and global Score and Noul answers.
- Decision frequency
- On each simulated dispatch wave.
- Fallback
- A deterministic local dispatcher runs without a key and supplies policy when answers are uncertain.
How it works
Overview
A browser simulation sends a shared world snapshot to the server for each dispatch wave. Jev returns a probability distribution for each specialist team plus global risk and reassessment judgments. Code blends uncertain answers with a local policy and solves a capacity-constrained assignment; it also runs a same-snapshot local counterfactual. The scenario is synthetic, and its recorded results do not establish a real emergency-response advantage.
- Language
- TypeScript
- License
- MIT
- Author
- Jason-Doyle
Sources
Related projects
JevGPT
Build a chatbot by repeatedly asking Jev to choose the next word.
What Jev decides
Which dictionary word should follow, or whether the reply is complete.
Jev Chat Assistant
An Android conversation copilot that uses Jev to interpret chats and rank reply candidates without sending for the user.
What Jev decides
The likely intent, conversation risk, reply need, best next action, unresolved tension, and which of three candidate replies fits best.
TypeSafe Mario
Jev plays Super Mario Bros. from structured emulator telemetry rather than screenshots.
What Jev decides
The next controller action, whether jumping is useful now, and the immediate danger level.