12-03, 13:30–14:00 (UTC), AI/ML Track
As organizations increasingly integrate and adopt AI and machine learning internally, the challenge of maintaining separate pipelines for ML-powered systems and conventional software makes it difficult for DevOps teams to maintain these separate pipelines. This talk explores a unified approach to DevOps and MLOps, demonstrating how existing DevOps pipelines can be transformed into efficient MLOps pipelines using ModelKits with KitOps
We'll begin by examining the reasons behind the traditional separation of DevOps and MLOps pipelines, including differences in project nature, required expertise, and the size and complexity of artifacts. We'll then delve into the challenges posed by separate pipelines, such as increased costs, coordination difficulties, and accumulating technical debt. Thus the attendees will learn how to leverage open source tooling like KitOps to create a unified pipeline that accommodates both traditional software and ML-powered projects, ultimately leading to more efficient and cost-effective operations.
Key topics include:
The unique challenges of ML projects in a DevOps context
Shortcomings of current MLOps pipelines
Introduction to ModelKits and their role in unifying pipelines
Step-by-step guide to transforming a DevOps pipeline into an MLOps pipeline
Practical examples of implementing ModelKits in existing workflows
This talk will explore how KitOps standardizes packaging through ModelKits, enabling seamless collaboration between data scientists and DevOps teams. By enhancing reproducibility, version control, and deployment automation
No previous knowledge expected
Shivay Lamba is a software developer specializing in DevOps, Machine Learning and Full Stack Development.
He is an Open Source Enthusiast and has been part of various programs like Google Code In and Google Summer of Code as a Mentor and has also been a MLH Fellow. He is actively involved in community work as well. He is a TensorflowJS SIG member, Mentor in OpenMined and CNCF Service Mesh Community, SODA Foundation and has given talks at various conferences like Github Satellite, Voice Global, Fossasia Tech Summit, TensorflowJS Show & Tell.