Building with Jev and System One Models Workshop: 3 Hands On Projects
Your pipeline asks hundreds of tiny questions a day.
You’re paying an LLM to answer every one.
Is this test result suspicious?
Which code issue matters most?
Is this AI feature worth its token bill?
These are decisions, not writing tasks. Sending each one to a large LLM makes it slow, expensive and unpredictable.
In this live workshop, you’ll replace those calls with Jev: a fast, structured decision model that returns a typed answer plus a confidence score your code can act on directly.
What you’ll build (all hands-on)
Project 1: Cut AI Costs: Measure, Route and Cascade
Track token cost, model usage and decision quality in an AI workflow. Then use Jev to send simple requests to a cheaper model and uncertain ones to a stronger model, and compare the results.
You’ll leave with: a reusable cost-routing pattern for your own pipeline.
Project 2: AI-Powered Exploratory Testing with Jev + Playwright
Connect Jev to a live app and Playwright to turn test observations into confidence-aware verdicts your automation can act on.
You’ll leave with: a workflow that flags suspicious behaviour without manual review.
Project 3: Build a Simple Agent with Jev
Build an agent where Jev makes the quick call at each step and hands only uncertain cases to an LLM.
You’ll leave with: a hybrid agent that’s faster and cheaper than an LLM-only one.
You’ll leave with
Working code for all three projects
A drop-in “decision layer” pattern for your own pipelines
A clear rule of thumb for when to use Jev and when to use an LLM
The recording and a certificate of completion
Who this is for
Software Developers building AI-powered workflows and applications
QA and Test Engineers exploring AI-assisted testing
DevOps and SRE Engineers working on triage, routing, and automation
AI Engineers designing hybrid Jev + LLM systems
Engineering Leaders looking for measurable AI efficiency rather than AI experimentation
Meet your instructor: Renata Andrade
Quality Engineering Lead · Test Automation Architect
Renata helps engineering teams build quality into how they work, in the easiest, fastest and most reliable way possible. She brings a tester's instinct to AI: don't just ask whether it works, ask whether you can trust the answer, and what it costs to find out.