skip to Main Content

Run open source LLM evaluations with Opik!

VINCENT KOC

Lead AI Researcher Engineer & Developer Relations at Comet

Vincent Koc is a globally recognized lecturer, futurist, and keynote speaker known for his work as an artificial intelligence engineer and technical leader. With over two decades of experience across industries such as finance, telecommunications, travel, and FMCG, he has led data-driven projects for major organizations including Qantas, McDonald’s, Cisco, and the Australian Federal Government. Vincent frequently presents at prestigious conferences worldwide and serves as a guest lecturer and mentor at universities in Australia and the U.S., sharing expertise in areas from data governance to generative AI. He was named one of IAPA’s Top 25 Analytics Leaders in 2020 and has received accolades from the Content Marketing Association and Content Marketing Institute for his impactful consumer-facing projects. A fellow at the Institute of Managers and Leaders Australia, Vincent is dedicated to advancing the field of data, mentoring future professionals, and informing the public about rapid technological change.

May 14th, 1:15 – 1:45PM ET

Precision Prompting: Tuning LLM Performance Through Algorithmic Optimization

In the era of foundation models and agentic systems, prompt engineering has emerged as a crucial skill—but one that remains a dark art and not well understood. This talk introduces a range of concept tied to precision prompting, a systematic approach to optimizing prompts using algorithmic techniques drawn from evolutionary computation, bayesian search, and LLM-guided reasoning. Showcasing the Opik Optimize SDK and learnings in building optimization algorithms, we’ll explore how in-the-loop evaluation, trace feedback, and structured optimization strategies can dramatically improve LLM performance across a range of tasks. You’ll walk away with a deeper understanding of how to move beyond handcrafted prompts and toward reproducible, data-driven, and production-grade tuned LLM prompts.

Back To Top