

MLOps & LLMOps with Noah Gift #38
13 snips Nov 30, 2023
In this discussion, Noah Gift, MLOps leader and executive in residence at Duke University, shares insights from his 30 years of experience, including building data pipelines in the film industry. He emphasizes the crucial role of MLOps and the software engineering skills essential for data scientists. Noah contrasts Python and Rust, advocating for flexibility in choosing tools. He delves into the differences between MLOps and LLMOps, discussing security concerns and the future of deployment strategies, making a compelling case for adapting to the tech landscape.
AI Snips
Chapters
Transcript
Episode notes
Early Film Industry Data Engineering
- Noah Gift worked in the film industry where data engineering began with digital workflows in early 2000s.
- He encountered lots of bad software practices but also interesting rapid creative coding tasks.
From Film to Startups to Teaching
- Noah left the film industry due to poor software quality and moved to startups where software mattered more.
- He learned startups are designed to fail and working in them was like paying to learn, before teaching full-time.
MLOps Needs DevOps Experience
- To do MLOps successfully, you must understand DevOps and software engineering fundamentals.
- Deploying models in production requires automation, observability, and production mindset beyond data science.