✨ Product Innovation with AI ✨ AI Evals in Practice

​How do teams go from prompt to reliable output?

​We’re back with the 4th edition of Product Innovation With AI - this time, we’re focusing on one of the hardest (and most overlooked) parts of building with AI: Evaluation.

​It’s easy to get a model to generate something.
It’s much harder to know if that output is actually good, consistent, and reliable enough to ship. The real challenge is building a robust, repeatable way to assess that output.

​This session is all about how teams are tackling that challenge in practice.

​You’ll hear from product and AI practitioners working hands-on with LLMs, sharing how they:

​↳ define what “good output” actually means
↳ design evaluation frameworks and workflows
↳ move from prompt experimentation to reliable, production-ready systems
↳ balance speed vs. quality when shipping AI features

​🎤 Speakers:
Builders who are actively turning AI from experiments into reliable product experiences — and will share how they evaluate what actually works:

​↳ Loredana Ninov - Senior Product Manager, Customer Engagement & Discovery, eMAG
↳ Mădălina Țurlea - Co-Founder, Lovelaice
↳ Albert Cristea - Product/Project Delivery Director, Clever Craft d.o.o.
↳ George Diță - Product | M&A Intelligence & AI Systems, Trilogy

​This session is designed for product managers, engineers, and builders integrating AI into their products or services - especially if you're already experimenting with prompts and now need structure, confidence, and reliability.

​📍In-person event at eMAG Headquarters

​ 📆 April 29th, 18:30 (Bucharest Time, RO)

​ ⏳ 2h of parallel interactive sessions & networking afterwards

​ 🎟️ Limited spots available