Documentation↗
Book a Demo

Best Practices: ML Model Management and Versioning

About

In this practical session, Seldon’s Customer Success Engineer Maciej Kozubal dives into the critical, but often overlooked, discipline of model management and versioning. Whether you’re building ML infrastructure, deploying models in production, or supporting compliance and rollback strategies, this talk outlines the real-world risks of poor version control, and how Seldon’s tools simplify this process.

Designed for both technical and business-focused audiences, the session walks through key challenges, live tooling examples, and the emerging regulatory pressures that make robust model management essential for any ML team.

Learnings

  • Why Model Versioning Matters
    Learn the business and technical risks of not tracking your models, including: compliance violations, model drift, biased outcomes, and rollbacks without audit trails.

  • Model Lifecycle Challenges in Production
    Explore the full picture of ML ops: from retraining and monitoring to drift detection, adversarial attacks, explainability, and versioning complexity across dozens of models.

  • Real-World Rollback and Audit Scenarios
    Understand how versioning supports incident response, legal compliance (e.g. EU AI Act), and post-deployment investigations when models misbehave.

  • Seldon’s Model Management Approach
    See how Seldon supports both semantic versioning (in Core 1) and numerically structured folders (in Core 2), with automated logging, rollback, and lineage tracking.

  • Multiple Deployment Methods
    Use GitOps, CLI, SDK, or the Seldon Enterprise UI to deploy, monitor, and manage models programmatically or visually.

  • Core vs. Core V2 Improvements
    Learn how Core v2 simplifies deployment with minimal manifests, one-line updates, and persistent endpoints—while enabling powerful version switching behind the scenes.

  • Compliance-Ready Tooling
    Explore how model versioning fits into responsible AI practices, including drift detection, explainability, model performance metrics, and inference auditing.

Picture of Maciej Kozubal

Maciej Kozubal

Maciej Kozubal, PhD is an ML/MLOps Solutions Architect at Seldon, specializing in architecting scalable ML and GenAI systems for enterprise production. With a unique background as a semiconductor physicist and particle accelerator operator, Maciej combines deep scientific rigor with practical engineering expertise. His work spans the full ML lifecycle from designing LLM/GenAI pipelines (RAG, vector search, embeddings), implementing explainability and drift detection, optimizing infrastructure for latency and cost, to guiding customers through deployment strategies on Kubernetes. Known for delivering impactful PoCs, running enablement sessions, and driving adoption across diverse industries, Maciej excels at translating complex AI capabilities into production-ready solutions that deliver business value.

Complex, real-time use cases is what we do best

Talk with a expert to explore how Seldon can support more streamlined deployments for real-time, complex projects like fraud detection, personalization, and so much more.

Book a Demo

MORE RECENT WEBINARS & EVENTS

Building Trust in Production ML: A Complete Guide to Model Observability
Production ML Serving & Monitoring with Seldon & Kubernetes
Europe’s Landmark Regulation: What You Should Know
The Foundation of AI-Enabled Apps
An Essential Guide to ML Model Serving Strategies (Including LLMs)
DeepSeek Reasoning, the Cake Mystery
Stay Ahead in MLOps with our
Monthly Newsletter!

Join over 25,000 MLOps professionals with Seldon’s MLOps Monthly Newsletter. Your source for industry insights, practical tips, and cutting-edge innovations to keep you informed and inspired. You can opt out anytime with just one click.

Email Signup Form