PyData London 2024

Are you ready for MLOps? 🫵
06-16, 15:00–15:40 (Europe/London), Salisbury

MLOps has survived the hype cycle and is gaining in maturity. But are we looking at MLOps for answers for the right things?
No matter how valuable MLOps can be for you, without proper building blocks in place MLOps cannot live up to its full potential. What are the prerequisites for MLOps? What parts of MLOps should you focus on? When should you even start thinking about MLOps, or when is ‘plain’ DevOps wiser to focus on first? Join us in this session to learn more!

Let's learn more about MLOps together! In this talk, we will share with you our take on when and why MLOps is relevant for you. So, you learn when and how to apply MLOps the right way to get most value out of it. Our collective experience entails multiple years of consulting companies on the topic, as well as teach trainings on MLOps. Having applied MLOps in many contexts and situations allowed us to see commonalities between them — and those we would like to share with you! Why? Because we love sharing knowledge!

Contents of the talk 📌

  1. [5 min] Intro
    1. Who are we
    2. MLOps: are you ready? Let’s find out.
  2. [8 min] The promise of MLOps (why MLOps)
    1. MLOps in the hype cycle
    2. What MLOps can offer you: Experiment tracking, Dataset versioning, Model versioning, Monitoring, ML pipelines, Model registry, CICD
    3. Breakdown: where do all these developments come from?
  3. [8 min] The ultimate inspiration source called DevOps
    1. How DevOps came and stayed
    2. What parts of MLOps are actually DevOps
    3. Diving deeper: what good software products consist of. About robust software products
    4. And what else?
  4. [6 min] The MLOps twist
    1. The organisational perspective
    2. Team enablement and autonomy
    3. Working together
  5. [4 min] Summing things up
    1. When to consider MLOps
    2. First things first: where to start
    3. Let’s together make this a great community 🫂
  6. [4 min] End

[35 minutes total / 5 min QA]

🏡 What you will take home

At the end of the talk, you will be taking home an understanding of:
- The current position- and maturity level of MLOps in the Data industry
- The main components of MLOps
- When MLOps is most valuable
- How MLOps was created: what ideas it lent and reused from other methodologies
- How DevOps and MLOps intermix and where you should focus on first
- How MLOps is unique compared to other software good practices and frameworks

🏡

❤️ Open Source Software

MLOps is presented in a cloud-agnostic way. Software libraries mentioned are open source. There is no agenda for representing any major cloud.

🎒 Pre-requisites

The talk can be followed without MLOps familiarity ✓. Recommended, though, is interest in MLOps ♡.

Prior Knowledge Expected –

Previous knowledge expected
Jetze Schuurmans

Jetze is a well-rounded Machine Learning Engineer who is as comfortable solving Data Science use cases as he is productionizing them in the cloud. His interests include MLOps, GenAI and Cloud Engineering. As a researcher, he has published papers in Computer Vision, Natural Language Processing and Machine Learning in general. Jetze loves sharing knowledge during community events and by giving trainings.
Jeroen Overschie

Jeroen is a Machine Learning Engineer at Xebia Data (formerly GoDataDriven), in The Netherlands. Jeroen has a background in Software Engineering and Data Science and helps companies take their Machine Learning solutions into production.
Besides his usual work, Jeroen has been active in the Open Source community. Jeroen published several PyPi modules, npm modules, and has contributed to several large open source projects (Hydra from Facebook and Emberfire from Google). Jeroen also authored two chrome extensions, which are published on the web store.

Hope to see you at PyData London 🇬🇧! 👋🏻