The role of CDEs in Generative AI
summary
Career story links storage and ML ops: built ORC file format that yields 20% storage savings and roughly a billion dollar impact, then built FB Learner Flow that serves every production model. Developers iterate by overfitting small datasets, scaling up, then hyperparameter sweeping; GPUs are scarce and costly so cloud dev environments must version code data and model weights together and support CICD for models. Notes practical tools and workflows like Gitpod, Copilot, ChatGPT copy‑paste patterns, and vector DBs as feature stores.
AI-written summary · about
chapters (8)
- 0:00CDEs and Generative AI
- 1:02Pam’s career journey GPUs to AI
- 6:20AI hype cycle inflection point
- 10:16AI impact on developer workflows
- 12:03Developer experience AI models
- 22:04Ideal CDE for AI development
- 27:20Q&A AI interfaces future
- 30:05Closing thoughts prototyping
related talks
49mGenerative AI: Concepts, Tools, and Applications for Data Professionals
27mTransforming Developer Platforms with Generative AI
36mUnleashing Generative AI: Improving Developer Productivity in SRE
47mLessons learned from using generative AI dev tools
15mBuild for Speed: Lessons from Empowering GenAI Research at Canva
28mSoftware 3.0 and The Emerging AI Developer Landscape
37mGenerative AI: Beyond (Just) Hype
19mDeveloper Centric AI