MLOps community #7 (in-person | en personne)

📣 And the fun continues with an all-new MLOps Community Meetup at the Computer Research Institute of Montreal (CRIM)! 🎉 🏢 On the docket:

• Thierry Jean, Machine Learning Engineer for Hamilton @DAGWorks
• Ahmed Moubtahij, ing., NLP Scientist/ML Engineer at the CRIM.

IMPORTANT :

• The access point is on 405 Avenue Ogilvy, bureau #101, Montréal (Parc Metro - Free parking)
• Get your proof of attendance with you (available on the meetup event page or received per email), for us to check your status (waiting list or not) at the venue

📊🤖 Don't miss this opportunity to explore cutting-edge applications and ignite your curiosity. Save the date and secure your spot now! 👉🗓️ See you there! 👋😄

AGENDA

• 17h30 - Open doors
• 18h00 - Introduction
• 18h10 - Talk #1
• 18h50 - Break, Pizza & Networking
• 19h10 - Talk #2
• 19h50 - Final notes, break and networking
• 21h00 - End of event

TALKS

1. Think micro: why modular machine learning pipelines are better, by Thierry Jean

Description of the talk
Which step in my pipeline failed? Why did we include this feature? Can I delete this code? What does this metric mean? Data pipelines involve a ton of methodological decisions, but most are left hidden in your pipeline. By starting micro and building a modular codebase, you'll facilitate maintainability and collaborative development. In this talk, I'll share hands-on tips from my journey building a pipeline to benchmark hundreds of models using Hamilton.

2. Guardrails for LLM systems, by Ahmed Moubtahij, ing.
Description of the talk
Ever had an automated chat response that made you raise an eyebrow? You are not alone in questioning the reliability of stochastic language models. What if there were guardian agents ensuring only compliant responses reach your users? Enter LLM Guardrails. In this presentation, we'll dive into the critical role of quality control agents within LLM systems. Discover how these guardrails evaluate LLM outputs against stringent criteria before exposing outputs to the user. We'll explore the possibilities of corrective action and escalation when outputs fail to meet standards, and how this not only enhances service quality but also mitigates reputational risk for the provider.

SPEAKER BIOS
Thierry Jean: Thierry is a machine learning engineer for the open-source Python library Hamilton. He previously conducted scientific research, did AI consulting, and taught data science. He's driven by building tools to improve data science ergonomics and create a top-quality developer experience.

Ahmed Moubtahij, ing.: Ahmed Moubtahij, ing., automation engineer by training, MSc.A., NLP scientist at CRIM. I've been passionate about NLP and professionally involved in it since 2020. Prior to that, I enjoyed 4 years of teaching C++ and Python. My free time goes to family, books of all sorts, podcasts, the twitterverse and open-source contributions.