MLOps for GenAI Applications - MLOps Podcast #256

Overview

Dive into a comprehensive podcast episode exploring MLOps for Generative AI applications with Harcharan Kabbay, Lead Machine Learning Engineer at World Wide Technology. Gain insights into the Retrieval-Augmented Generation (RAG) framework and its integration with MLOps best practices. Learn about automating platform provisioning, application design principles, and the use of Kubernetes in AI systems. Discover strategies for reducing development time, enhancing security, and implementing effective monitoring in AI applications. Explore topics such as CI/CD pipelines, version control, and automated deployment processes for maintaining agility and efficiency in AI projects. Benefit from Kabbay's expertise in MLOps, DevOps, and automation as he shares valuable insights on building scalable, automated AI systems and integrating RAG-based applications into production environments.

Syllabus

[] Harcharan's preferred coffee
[] Takeaways
[] Against local LLMs
[] Creating bad habits
[] Operationalizing RAG from CICD perspective
[] Kubernetes vs LLM Deployment
[] Tool preferences in ML
[] DevOps perspective of deployment
[] Terraform Licensing Controversy
[] PR Review Template Guidance
[] People processes tech order
[] Register for the Data Engineering for AI/ML Conference now!
[] ML monitoring strategies explained
[] Serverless vs Overprovisioning
[] Model SLA's and Monitoring
[] LLM to App transition
[] Ensuring Robust Architecture
[] Chaos engineering in ML
[] Wrap up