Setting up Jev routing in 10 minutes to pick the best-value model per prompt
A tutorial walks through 10 steps from creating an API key to wiring up a Cursor agent so Jev picks a model by cost, balanced or intelligence tier, claiming about 80% savings in cost and time.
I found the "Internet" moment in using Jev Jev analyzed every LLM and chose the BEST-VALUE one for each prompt It cut my costs and time by ~80% - and it’s probably the best way to use Jev... here's how to setup it in 10 min: step 1 → open @typesafeai , create API key (keep it off chat paste) step 2 → store TYPESAFE_API_KEY in a local .env step 3 → clone the Github below, then npm install && npm test && npm run demo step 4 → test Jev before connecting an agent: npx tsx sdk-runner/cli.ts --route-only "Rename a button label" step 5 → dry-run first: npx tsx sdk-runner/cli.ts --dry-run "your task" - read tier + model before you spend step 6 → add CURSOR_API_KEY + npm i cursor/sdk → live: npx tsx sdk-runner/cli.ts --run "your task" step 7 → Jev picks cost / balanced / intelligence → that model runs → you get the result step 8 → financial / production / publish / send-email stop unless --approve step 9 → kill switch: enabled: false or bypass step 10 → use it! the result: your agent stops wasting maximum intelligence on easy work - Jev picks the cheapest model that can win, then unlocks maximum power only when failure gets expensive unfortunately, AutoMode works extremely poorly across all AI tools and wastes even more tokens - this setup finally solves that problem Steal the agent setup everyone will use by 2028, copy repo - then read the complete Jev playbook below ↓
This post quotes another post: “Long read: how to build a Jev-powered agent stack”