Engineer your company’s future with AI: Use open source tools on AWS
summary
A fast overview of running open source MLOps on AWS with concrete examples: a camera+model system that saved a crayfish farm, starting on an Ubuntu laptop with Jupyter then scaling to cloud; recommended tools include Kubeflow and MLflow and advice to reuse on‑prem GPUs (many shops run at 20–30% GPU utilization). Covers hybrid/multi‑cloud patterns, how to avoid lock‑in, cost and security tradeoffs, and when simple predictive models beat generative AI.
AI-written summary · about
chapters (11)
- 0:00Intro
- 1:18How I got introduced to the public cloud
- 2:28Challenges in AI
- 3:42Use cases
- 4:45Benefits of AI
- 5:22What is open source
- 6:40Enterprise Readiness
- 7:19How do you scale
- 9:31Running AI on any cloud
- 11:31What are your AI projects
- 12:50Learn more
related talks
41mLeveraging open-source and 3rd party tools for ML on AWS
37mOptimizing AI and ML with Platform Engineering
16mKeynote: Accelerating AI/ML Development with Open Source Software
26mKubernetes and MLOps for Scalable and Reproducible Generative AI
29mAI/MLOps for Busy People: A Field Guide To Implementing Cloud Native AI/ML- 16mKeynote: Accelerating AI/ML Development with Open Source Software
37mStreamlining Machine Learning Operations
29mTeam Teardown: MLOps in the Cloud for IoT