Keynote: Noah Goodman, Uber AI Labs, Pyro, Stanford University in conversation with Jim Zemlin
Interviewat Open Source Leadership Summit 2019 · on Tuesday, 2019-03-12· at 10:00· The Ritz-Carlton Ballroom·
summary
Argues AI is early but converging: frameworks share interfaces (numpy and PyTorch), and progress comes from tooling not new algorithms — continuous integration and rigorous testing are the single biggest gains for production ML. Cites concrete stacks: Pyro built on PyTorch (Facebook) and probabilistic languages Church and WebPPL; predicts meaningful stabilization within five years while warning about data hoarding and advocating open data license style sharing to unlock network effects.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:55State of AI ML
- 3:58Two sides of AI
- 5:47The role of opensource
- 8:33The impact of AI
- 11:28Why is AI happening now
- 16:00The future of data
- 18:43Future of AI
related talks
12mKeynote: What's New in AI
26mKeynote: AI in the Real World: Today and Tomorrow
16mKeynote: An Open-Source Stack for AI Compute: Kubernetes + Ray + PyTorch + vLLM
17mKeynote: Machine Learning & A.I. At Uber
15mKeynote: Building an Open Agentic Future
16mKeynote: Open Source for AI and AI for Open Source
32mKeynote: Fireside Chat with Soumith Chintala, Co-Creator of PyTorch & Merve Noyan
32mKeynote: AI & the stuff built for AI - are they actually useful for data science?