Super Data Science: ML & AI Podcast with Jon Krohn 931: Boost Your Profits with Mathematical Optimization, feat. Jerry Yurchisin
20 snips
Oct 14, 2025 In this engaging discussion, Jerry Yurchisin, Senior Data Science Strategist at Gurobi, unveils the fascinating world of mathematical optimization. He explains how it differentiates from machine learning and shares intriguing use cases from Toyota and Total Wine. Listeners learn about practical games that illustrate optimization challenges and discover resources to get started with Gurobi. Jerry also touches on the future of quantum computing in optimization, emphasizing the importance of beginning to implement these strategies now for immediate business benefits.
AI Snips
Chapters
Books
Transcript
Episode notes
Optimization Produces Actionable Decisions
- Mathematical optimization defines decision variables, constraints, and an objective to produce actionable decisions.
- It finds optimal actions for complex systems where predictions alone aren't sufficient.
Play The Burrito Optimization Game
- Try the Burrito Optimization Game to experience optimization complexity firsthand.
- Compare your intuition to Gurobi's solver to see how human heuristics often fall short.
GPUs Help Large Linear Programs With New Algorithms
- GPUs can accelerate very large linear programs, but traditional pivot-based algorithms poorly map to GPU parallelism.
- Gurobi explores GPU algorithms and partners with NVIDIA for super-large LPs.
