Product Market Fit

Product Market Fit

AI Agent Templates

Jev: The Complete Founder’s Guide

Everything You Need to Know About the AI Model Built Only to Make Decisions, and How to Put It to Work in Your Startup

Guillermo Flor's avatar
Guillermo Flor
Sep 24, 2026
∙ Paid

On September 15, TypeSafe AI released Jev, a model built to make decisions instead of writing answers.

Its founder? Diogo Almeida, previously worked at OpenAI on research behind the methods that led to ChatGPT.

Give Jev a piece of text and a set of options, and it returns a probability for each one.

Is an email urgent? Which team should handle a support ticket? Does a lead fit your target customer? These decisions can take a fraction of a second and cost far less than asking a conventional LLM to write an answer.

That sounds narrow until you look at how companies use AI. Many calls only need a choice that software can act on. If those choices become cheap enough to make on every lead, ticket and agent action, startups can automate checks they previously had to skip.

X avatar for @CompleteSkeptic
Diogo Almeida@CompleteSkeptic
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x …
6:17 PM · Sep 15, 2026 · 39.5M Views

4.03K Replies · 8.22K Reposts · 75.9K Likes

📚 In this guide, you’ll have Jev answering real questions about your business, find where it could fit into your workflows and learn when it still needs a human check.

Inside you will find:

  1. How Jev Makes a Decision Without Writing an Answer

    See how Jev turns text into a choice your software can use.

  2. How to Run Your First Decision in Five Minutes

    Test it on an email, a sales lead, and a support ticket: no code needed.

  3. Which Decisions in Your Startup Should Move to Jev

    Audit your workflows and identify the first decisions worth testing.

  4. How to Cut Your Coding Agent Bill by 10x

    Shrink Claude Code sessions by 10x and route each task to the cheapest model.

  5. How to Read 100,000 Customer Conversations for Under a Dollar

    Find recurring problems, churn risks, and feature requests across all your customer conversations.

  6. Where Jev Gets Decisions Wrong

    Learn how to read its uncertainty and decide when a human should step in.


🫸🏻 Wait a second. New Founder Perks just landed!

Premium members can now access:

  • Pitch Deck Review (Founding members): submit your deck and get structured feedback on narrative, positioning, market framing, traction, business model and fundraising ask, before it reaches investors.

  • Startup Spotlight (Annual and Founding members): we feature your startup in a banner inside Product Market Fit, in front of 80,000+ founders, investors and operators.

They come on top of everything annual premium subscribers already unlock: the full library (200+ decks, 10,000+ investors, AI Agent playbooks) and more than $3,000 in exclusive deals on the AI tools founders use every day.

Framer, Granola, Gojiberry, CodeWords, Throxy, Omnia, MITO, Slashy, Cubic, QuiverAI, LightSprint, Hera and many others! We add new deals every week.

Already a premium subscriber? Unlock your perks here! Not yet?

Upgrade Here!

The rest of the issue is available exclusively to paid subscribers!

Let’s begin!


1. How Jev Makes a Decision Without Writing an Answer

Keep reading with a 7-day free trial

Subscribe to Product Market Fit to keep reading this post and get 7 days of free access to the full post archives.

Already a paid subscriber? Sign in
© 2026 Guillermo Flor · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture