Course curriculum

    1. Introduction

    2. What is Model Registry?

    3. Logging Models

    4. Linking Models to Registry

    5. Navigating Model Registry

    6. Quick question before you continue

    1. Introduction to Automations

    2. Webhooks

    3. Hosting a Webhook Server

    4. Webhooks in Weights & Biases

    5. Testing Webhooks

    6. Creating a Webhook Automation

    7. Webhook Exercises

    1. LLM Case Study Overview

    2. Introduction to Launch

    3. Finetuning an LLM and Saving Model

    4. Setting Up LLM Evaluation

    5. Setting Up Launch

    6. Creating a Launch Automation

    7. LLM Evaluation Results

    8. Model Management Exercises

    1. Automation Design Patterns

    2. Enterprise Model Management Features

    1. Create an Automation

    2. Course Certificate

About this course

  • Free
  • 25 lessons
  • 2.5 hours of video content

Your instructors

Hamel Husain

Founder @ Parlance Labs

Hamel is currently a founder at Parlance Labs, a research lab and consultancy focused on LLMs. Previously he was an entrepreneur in residence at fast.ai, where he built tools for data scientists and machine learning engineers. He also worked at Airbnb, DataRobot, and GitHub where he built a wide array of machine learning products and infrastructure. Hamel has contributed to data and infrastructure tools in open source such as Metaflow, Kubeflow, Jupyter, and Great Expectations. Hamel was also a consultant for over 10 years, and used data science to improve business outcomes in the restaurant, entertainment, telecommunications, and retail industries.

Noa Schwartz

Product Manager @ Weights & Biases

Noa Schwartz is a Product Manager at Weights & Biases, where she leads product for W&B Artifacts, Registries, and Automations and has worked with W&B's largest enterprise customers across every ML use case to streamline and add reproducibility to model lifecycle management. She previously worked at Scale AI as a PM on the Synthetic Data team, supporting the launch of Forge, Scale's first GenAI application in the e-commerce space. Noa earned her BS in Computer Science at MIT.

Darek Kłeczek

Machine Learning Engineer

Darek Kłeczek is a Machine Learning Engineer at Weights & Biases, where he leads the W&B education program. Previously, he applied machine learning across supply chain, manufacturing, legal, and commercial use cases. He also worked on operationalizing machine learning at P&G. Darek contributed the first Polish versions of BERT and GPT language models and is a Kaggle Competition Grandmaster.

What pain points does a Model Registry solve for me?

I am a...

  • MLOps Engineer

    I need a source of truth for the latest and greatest models the practitioners need me to deploy and an easy way to consume them.

  • MLE

    I need to publish my new best model versions, easily compare new candidate models to the current baseline or production model, and handoff my model for deployment to an MLOps Engineer.

  • ML Team Lead or Executive

    I need to see all the ML tasks my team is creating models for and track which models are in production and how they are being iterated over.

  • Part of the Compliance team

    I need to enforce data governance policies, track data sources used for training, and ensure compliance with data privacy regulations.

  • Product Manager

    I need documentation for the models the team is working on, their expected inputs and outputs, and up-to-date performance results.

Your Goals

Complete this course to:

  • Create an enterprise-level Model Management System

    You will understand and navigate the complexities of model versioning and governance at an enterprise level, ensuring models are managed with precision and accountability.

  • Automate workflows with Webhooks

    You will be able to use model registry webhooks and triggers for workflow automation, streamlining the testing, deployment and management of machine learning models.

  • Evaluate & improve models with metrics and evaluation techniques

    Through detailed case studies and hands-on exercises, you will improve in conducting thorough model evaluations, ensuring they pass any enterprise benchmarks for accuracy and reliability.

Industry Leaders on Model Registry

“Having, dashboards across the office where you can see how things are moving, see how the models converging both in test, but also whenever it's deployed in production and seeing the impact on the fleet and communicating that out to everybody else in the company and hearing all of our partner nerds who nerd on different things celebrate that success. I think it just really becomes an interesting virtuous cycle. ”

Angela Manzo, Head of Analytics, Data Science and ML @ iRobot

“The W&B Model Registry simplifies our lives in so many ways. It brings less noise to the user experience, as we are now only seeing models that are production-ready. It stores all the production-level information we need.”

Thibault Main de Boissiere, ML Platform Team Lead @ Canva

“We use the model registry as our source of truth. It’s a key component of our backend.”

Nantas Nardelli, Senior Research Scientist @ Carbon Re

“Using the W&B Model Registry for model promotion has been extremely useful. Having the full history of the model available also makes it very easy for other departments — like product management — to help manage the deployment process.”

Benedict Eugine, Data Scientist @ M-KOPA

Here is what people had to say about Hamel's previous courses

5 star rating

Insightful and concisely explained course

Simran Simran

I have found the course really informative and helpful. With this course, I came to understand the concepts behind CI/CD, GitHub action workflows and how to leverage them for MLOps effectively. The explanation was also comprehensive and clear. I'd...

Read More

I have found the course really informative and helpful. With this course, I came to understand the concepts behind CI/CD, GitHub action workflows and how to leverage them for MLOps effectively. The explanation was also comprehensive and clear. I'd like to thank the course instructor for the same. With this course, I learnt about the paradigms of GitOps, wandb api and code testing which I'm sure will help in building and deploying ML pipelines and tracking experimentations. Looking forward to more such intriguing courses on wandb platform!

Read Less
5 star rating

Great content

Ama Ama

Really great content and worth going over it twice to solidify concepts

Really great content and worth going over it twice to solidify concepts

Read Less
5 star rating

Excellent course!

Rashmi Banthia

Very comprehensive, Thank you!

Very comprehensive, Thank you!

Read Less
5 star rating

Great Course!

Vincent Tu

I'm a little biased because I use W&B regularly, but this was a well-made course! Even though I was familiar with wandb Reports and Git/GitHub/Actions, I still learned a ton. I learned more about the Reports SDK and ghapi!

I'm a little biased because I use W&B regularly, but this was a well-made course! Even though I was familiar with wandb Reports and Git/GitHub/Actions, I still learned a ton. I learned more about the Reports SDK and ghapi!

Read Less
5 star rating

A great course

Patrik Reizinger

Read Less

Recommended prerequisites

  • Basic knowledge of machine learning

  • Familiarity with Python programming

Model CI/CD

Start automating your model registry and evals today.

Get started