PyData Nights Vol.2 MLOps Edition

Hallo Münchners,

This time Marty is back to MLOps land of the future!!!

We would like to invite you to our meetup with two exciting talks at a very cool location, for which we want to thank JetBrains for hosting us this evening along with drinks and pizza.

This event is brought to you in collaboration with the Munich🥨NLP community. Join their Discord to discuss the latest developments and also stimulate exchange on research and innovation around NLP.

Hurry up we have limited spots and see you on the other side.

Best,
Muhtasham and Nithish

=== Agenda ===
18:30 - Doors open
19:00 - Welcome and Introduction
19:10 - Talk 1: Reproducible ML workflows & dev environments with dstack by Andrey Cheptsov, dstack
19:50 - [Break] Networking and refreshment
20:00 - Talk 2: Why ML should be written as pipelines from the get-go by Hamza Tahir, ZenML
20:40 - Networking with Drinks

=== Talks ===
Talk 1: Reproducible ML workflows & dev environments with dstack
Speaker: Andrey Cheptsov, dstack

Abstract:
Building ML models is an iterative process. Let’s talk about tools and practices that help set up your dev environments and workflows for better productivity and reproducibility?
As a bonus, we’ll have an overview of dstack, an open-source utility that simplifies the MLOps stack, and helps run ML workflows and dev environments in the cloud.

Speaker Bio
Andrey is the creator of dstack. He is passionate about open-source and developer tools for AI. Previously, Andrey worked at JetBrains with the PyCharm team.

Talk 2: Why ML should be written as pipelines from the get-go
Speaker: Hamza Tahir, ZenML
Abstract: The mechanism through which ML propagates through an organisation from experimentation to production is key to its success. Oftentimes, there is a tendency to break this mechanism into a multi-step process, where experimentation workflows are siloed from their production counter-parts. This "Throw it over the wall" anti-pattern can stunt the velocity of ML teams. In this talk, we talk about why teams should unify this multi-stage process, and give data scientists more agency to exercise control over their production workflows. We'll also go through a practical demonstration with creating a unified MLOps pipeline with ZenML

Speaker Bio
Hamza Tahir is a software developer turned ML engineer. An indie hacker by heart, he loves ideating, implementing, and launching data-driven products. His previous projects include PicHance, Scrilys, BudgetML, and you-tldr. Based on his learnings from deploying ML in production for predictive maintenance use-cases in his previous startup, he co-created ZenML, an open-source MLOps framework to build portable production-ready ML pipelines.