MLOps.community MLOps Critiques // Matthijs Brouns // MLOps Coffee Sessions #100
MLOps Coffee Sessions #100 with Matthijs Brouns, MLOps Critiques co-hosted by David Aponte.
// Abstract
MLOps is too tool-driven; don't let FOMO drive you to pick the latest feature/model/evaluation/ store, but pay closer attention to what you actually need to release more safely and reliably.
// Bio
Matthijs is a Machine Learning Engineer, active in Amsterdam, The Netherlands. His current work involves training MLEs at Xccelerated.io. This means Matthijs divides his time between building new training materials and exercises, giving live trainings, and acting as a sparring partner for the Xccelerators at their partner firms, as well as doing some consulting work on the side.
Matthijs spent a fair amount of time contributing to their open scientific computing ecosystem through various means. He maintains open source packages (scikit-lego, seers) as well as co-chairs the PyData Amsterdam conference and meetup.
// MLOps Jobs board
// Related Links
https://www.youtube.com/watch?v=appLxcMLT9Y
https://www.youtube.com/watch?v=Z1Al4I4Os_A
--------------- ✌️Connect With Us ✌️ -------------
Join our Slack community: https://go.mlops.community/slack
Follow us on Twitter: @mlopscommunity
Sign up for the next meetup: https://go.mlops.community/register
Catch all episodes, blogs, newsletters, and more: https://mlops.community/
Connect with Demetrios on LinkedIn: https://www.linkedin.com/in/dpbrinkm/
Connect with David on LinkedIn: https://www.linkedin.com/in/aponteanalytics/
Connect with Matthijs on LinkedIn: https://www.linkedin.com/in/mbrouns/
Timestamps:
[00:00] Introduction to Matthijs Brouns
[00:28] Takeaways
[03:09] Best of Slack Newsletter
[03:38] AI MLFlow
[04:43] Nanny ML
[05:08] Best confinement buy over the last 2 years
[06:35] Matthijs' day-to-day
[08:24] What's hot right now?
[09:36] ML space, orchestration, deployment
[10:21] Scaling
[13:20] Low-risk releases
[15:27] Scale Limitations or Fundamental in API
[16:33] MLOps maturity to a certain point
[18:57] Interdisciplinary leverage needed
[21:11] PyScript
[22:41] Next pipeline tools
[24:02] General pattern to build your own tools
[30:25] Technology recommendation to a chaotic space
[33:46] Structured data vs tabular data
[35:52] Big barriers in production
[37:57] Standardization
[39:20] Automation tension between the engineering side and the data science side
[41:50] Low-hanging fruit
[42:30] Human check
[43:43] Rapid-fire questions
[48:30] PyData Meetups