In this episode, Valerii Babushkin, Senior Director at BP and former Meta E7, shares insights on career transitions, ML system design, and navigating high-level IC and management roles. He discusses why he moved between IC and leadership positions, the challenges of onboarding at top tech companies, and how to identify what truly matters in a company. Valerii also dives into engineering communication, the importance of documentation, and why a strong writing culture is key for success. We also explore ML system design principles, including how to write an effective design doc, choosing the right loss functions and metrics, and avoiding common pitfalls like data leakage. Valerii shares his thoughts on Kaggle, when it’s useful for career growth, and what differentiates staff, senior staff, and principal engineers. If you’re aiming to level up in ML, this episode is packed with actionable career and technical advice! 🚀 🔗 Resources Mentioned 📖 Valerii’s Book: Machine Learning System Design https://amzn.to/4hGaX5l 🔗 Connect with Valerii on LinkedIn: https://www.linkedin.com/in/venheads 📄 Staff Engineer Book: https://amzn.to/4hCfq9l

In this episode, we sit down with Ville Tuulos, CEO of Outer Bounds and creator of Metaflow, to discuss why machine learning is so hard—and how to make it easier. Ville shares his journey from training neural networks at 13 to leading ML infrastructure at Netflix, where he built tools to empower data scientists. We dive into ML cycles, user empathy, infrastructure, and what separates great ML engineers from the rest. 📚 Learn More About Ville Tuulos & Metaflow - Ville Tuulos on LinkedIn: https://www.linkedin.com/in/villetuulos/ - Outer Bounds: https://outerbounds.com - Effective Data Science Infrastructure: How to Make Data Scientists Productive (Ville’s book): https://amzn.to/4hN0Yee (affiliate link)

Dr. Kevin S. Van Horn joins the podcast to discuss Bayesian probability, ML optimization, and lessons from his career. He shares insights on NP-hard problems in ML, Bayesian methods, and how to ensure success in ML projects.

Kevin’s Newsletter: https://epistemicprobability.substack.com/
Jaynes' Book on Bayesian Probability: https://amzn.to/42GFOKu