Super Data Science: ML & AI Podcast with Jon Krohn 987: AI Infrastructure, Ray, and Why Nonlinear Careers Win, with Linda Haviv
34 snips
Apr 28, 2026 
Linda Haviv, AI infrastructure practitioner and former AnyScale developer advocate who also creates tech education content. She talks about staying current in AI, why system thinking may outgrow pure coding, the power of nonlinear career paths and side projects, and how open source is closing the gap with proprietary AI models.
AI Snips
Chapters
Books
Transcript
Episode notes
Use Side Projects As Career Insurance
- Do build a personal brand and side projects because they create career insurance and open unexpected opportunities like job offers or consulting gigs.
- Linda got hired at AWS after posting cloud content on TikTok and used side work to attract sponsors like Anthropic and consulting clients.
Publish Early To Accelerate Learning
- Do create content and teach in public even if small audience; the pressure to publish accelerates learning and invites feedback.
- Linda started with short music-plus-text videos while parenting and used incremental posts to build community and credibility.
Ray Delivers Python-Native Distributed AI
- Ray is a Python-native open source distributed computing framework tailored to AI workloads for training, data processing, and serving across GPUs.
- It evolved from reinforcement learning research at Berkeley and provides RayData, RayTrain, and RayServe so engineers avoid deep distributed-systems work.
