Jev: 8 real use cases for the fastest, cheapest model I’ve ever used | John Lindquist

<p><strong>John Lindquist</strong> created egghead.io, a developer education platform used by hundreds of thousands of working engineers. These days he’s building mega.dev, a hands-on program specifically for developers who want to do real work with AI agents, not just prototype them.</p><p><br></p><p><strong>What you’ll learn:</strong></p><ol><li>Why Jev is a decision engine, not a chatbot, and what that distinction actually changes about how you build</li><li>How John built a real-time voice to-do app that classifies and executes commands with no visible pause</li><li>The data deduplication pattern that merges messy records in milliseconds using confidence scores</li><li>Why Jev works best as a router, and how a single text input can navigate users deep into an app</li><li>What a chess match between Jev and a low-reasoning LLM reveals about speed, cost, and when to use which</li><li>The multi-step classification pattern John reaches for when one Jev pass isn’t enough</li><li>Where Jev falls short, and when you should still reach for a full generative model</li></ol><p>—</p><p><strong>Brought to you by:</strong></p><p><a href="https://www.vanta.com/howiai">Vanta</a>—Automate compliance and simplify security</p><p>—</p><p><strong>In this episode, we cover:</strong></p><p>(00:00) John Lindquist returns for Jev week</p><p>(04:32) What Jev actually outputs</p><p>(06:15) Demo: real-time voice to-do app</p><p>(08:17) How sequential Jev calls chain together</p><p>(10:38) Demo: plain English to function name (grocery cart)</p><p>(11:50) Demo: data deduplication and record merging</p><p>(13:45) Confidence scores and multi-model validation</p><p>(15:06) Demo: Jev as a multi-level app router</p><p>(18:23) Architecting around Jev</p><p>(19:35) Demo: Jev vs. traditional LLM at chess (speed and cost benchmarks)</p><p>(24:29) DOM interactions as a decision set, not an infinite canvas</p><p>(28:21) Demo: Wikipedia “path to philosophy” route mapper</p><p>(30:28) Demo: multi-agent coordination and collision avoidance</p><p>(33:36) Demo: real-time presentation coach</p><p>(36:56) Quick recap</p><p>(39:54) Lightning round and final thoughts</p><p>—</p><p><strong>Tools referenced:</strong></p><p>• Jev (TypeSafe AI decision model): <a href="https://typesafe.ai/blog/introducing-system-one-models-and-jev">https://typesafe.ai/blog/introducing-system-one-models-and-jev</a></p><p>• Vercel AI Gateway: <a href="https://vercel.com/docs/ai-gateway">https://vercel.com/docs/ai-gateway</a></p><p>• OpenRouter: <a href="https://openrouter.ai/">https://openrouter.ai</a></p><p>• Opus 5.5 (mentioned in context of iterative demo building): <a href="https://www.anthropic.com/claude-opus-5-5">https://www.anthropic.com/claude-opus-5-5</a></p><p>—</p><p><strong>Where to find John Lindquist:</strong></p><p>LinkedIn: <a href="http://linkedin.com/in/john-lindquist-84230766">linkedin.com/in/john-lindquist-84230766</a></p><p>X: <a href="https://x.com/johnlindquist">https://x.com/johnlindquist</a></p><p>Mega.dev: <a href="https://mega.dev/">https://mega.dev/</a></p><p>Egghead.io: <a href="https://egghead.io/">https://egghead.io/</a></p><p>—</p><p><strong>Where to find Claire Vo:</strong></p><p>ChatPRD: <a href="https://www.chatprd.ai/">https://www.chatprd.ai/</a></p><p>Website: <a href="https://clairevo.com/">https://clairevo.com/</a></p><p>LinkedIn: <a href="https://www.linkedin.com/in/clairevo/">https://www.linkedin.com/in/clairevo/</a></p><p>X: <a href="https://x.com/clairevo">https://x.com/clairevo</a></p><p>—</p><p>Production and marketing by <a href="https://penname.co/">https://penname.co/</a>. For inquiries about sponsoring the podcast, email jordan@penname.co.</p>