​About the event

​Stop paying an LLM to answer yes/no questions. Build three real decision layers with Jev in two hours.

​Your pipeline asks hundreds of small questions every day. Is this test result suspicious? Which code issue matters most? Is this AI feature worth its token bill? Most teams throw a large LLM at every one. There's a faster way.

​Not every AI problem needs another LLM call

​Generative models are great at writing, but they're slow, costly and unpredictable when all you need is a decision your code can act on.

​Jev is built for exactly that. You give it state, ask a typed question, and get back a structured decision with a confidence score. Your application can branch, rank or route on that answer directly. Think of it as a fast "System One" model that works alongside your LLMs rather than replacing them.

​This workshop takes Jev beyond the API demo. You'll build three different projects and see how the same decision primitives solve very different engineering problems.

​What you'll build

​Project 1: Exploratory testing with Jev + Playwright (hands-on)
Point Jev at a live application through Playwright and turn what it observes into confidence-aware verdicts. You'll get a test flow that tells you when something looks off and when to escalate, instead of a wall of brittle assertions.

​Project 2: Code health triage with RepoWise / CodeScene
Feed in static-analysis findings and let Jev classify, score and prioritise them. You'll get a decision layer that shows your team what deserves engineering attention first, so nobody has to review every issue by hand.

​Project 3: AI observability
Measure token cost, model usage and real impact, then use Jev to judge whether an AI workflow is earning its keep. You'll get a way to back "is this worth it?" conversations with data.

​Agenda (2 hours, live)

• ​0:00 – 0:30: What Jev is, its architecture, how it compares with LLMs, and Jev vs Laya

• ​0:30 – 1:00: Project 1, exploratory testing with Jev + Playwright

• ​1:00 – 1:25: Project 2, code health triage

• ​1:25 – 2:00: Project 3, AI observability, plus live Q&A

​You'll leave with

• ​Working code for all three projects

• ​A reusable "decision layer" pattern you can drop into your own pipelines

• ​A clear rule of thumb for when to use Jev and when to reach for an LLM

• ​Workshop recording

• ​Certificate of completion

​Who should attend

• ​QA engineers, SDETs and test automation engineers

• ​Engineering leads tired of drowning in tech-debt findings

• ​Platform and DevEx engineers building smarter CI pipelines

• ​AI engineers who want to cut LLM cost and latency

• ​Anyone who has asked "do we really need GPT for this?"

​Meet your instructor

​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.

​Renata Andrade
Quality Engineer Lead · Test Automation Engineer & Architect · AI Expert · Speaker · Instructor