Paul Iusztin
Senior AI Engineer / Founder at Decoding ML
Paul Iusztin is a senior AI/ML engineer with over seven years of experience building GenAI, Computer Vision and MLOps solutions. His latest contribution was at Metaphysic, where he was one of the core AI engineers who took large GPU-heavy models to production. He previously worked at CoreAI, Everseen, and Continental. In his last university year, he also tried building his first tech company, Dorel. He is the co-author of the LLM Engineer's Handbook, a bestseller on Amazon, which presents a hands-on framework for building LLM applications. Paul is the Founder of Decoding ML, an educational channel on production-grade AI that provides code, posts, articles, and courses, inspiring others to build real-world AI systems. Through Decoding ML, he collaborated with companies such as MongoDB, Comet, Qdrant, ZenML and 11 other AI companies. Paul's teaching career started by teaching the foundations of AI laboratory at the Politehnica University of Timisoara.
All Sessions by Paul Iusztin
Virtual Only: LLM & RAG Evaluation Playbook for Production Apps
<span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">Intermediate</span> </span>This workshop addresses the challenge of moving LLM/RAG applications from proof-of-concept to production. It emphasizes the critical need for evaluation and monitoring to optimize accuracy, latency, and costs. Participants will learn to quantify key performance metrics like latency, accuracy, and hallucinations. The workshop will cover adding a prompt monitoring layer, evaluating the retrieval step in RAG, and using LLM-as-judges for application-level metrics. Finally, attendees will learn to log metrics to a prompt management tool for effective experiment comparison and iterative improvement.
Virtual Only: LLM & RAG Evaluation Playbook for Production Apps
<span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">Intermediate</span> </span>This workshop addresses the challenge of moving LLM/RAG applications from proof-of-concept to production. It emphasizes the critical need for evaluation and monitoring to optimize accuracy, latency, and costs. Participants will learn to quantify key performance metrics like latency, accuracy, and hallucinations. The workshop will cover adding a prompt monitoring layer, evaluating the retrieval step in RAG, and using LLM-as-judges for application-level metrics. Finally, attendees will learn to log metrics to a prompt management tool for effective experiment comparison and iterative improvement.
