PyLadies Paris Tech Talks #22

Everyone 💚🐍 is welcome to our next on-site event 30th of September featuring
𓆙 Ingy ElSayed-Aly and Mariia Vladimirova:
and continuing with
⚡ lightning talks where you can take 3 mins to talk about anything Python or tech related (more below)

🌟Agenda

18h30 - 18h45 Come and take your seat

18h45 - 19h00 Welcome by PyLadies Paris and Criteo

19h00 - 19h30 Introduction to Reinforcement Learning by Ingy ElSayed-Aly

19h30 - 20h00 Bias in LLMs: what our tests can miss by Mariia Vladimirova

20h00 - 20h15 Lightning talks

20h15 - 21h30 Snacks and Networking

🌟 Ingy ElSayed-Aly
Talk Title: Introduction to Reinforcement Learning
Abstract: Reinforcement learning has quietly powered robotics, recommendation systems, and game-playing agents for years. RLHF and reward-based training are what turn raw language models into helpful assistants, and the latest reasoning models lean on RL to learn how to think step by step, not just imitate text. In this talk, we'll cover the basic mechanism behind RL (agents, environments, rewards, and how they learn from trial and error), then see it in action through live demos. No prior RL experience needed.

About Igny: Ingy is a Research Engineer for Saint-Gobain Recherche. Her area of expertise is at the intersection of Reinforcement Learning and Formal Methods. She is passionate about Reinforcement Learning, Safety, Games and Horses.

🌟 Mariia Vladimirova
Talk Title: Bias in LLMs: what our tests can miss
Abstract: Large language models (LLMs) are getting better at answering questions. But does that mean they are becoming less biased?
This talk brings together our two recent studies to explore how we test LLMs and what those tests can miss. We’ll see how changing a prompt can change our conclusions about fairness, and how models can favor answers from their own “team”, even as they get better at resisting wrong advice.
Through simple examples, we’ll explore why better scores don’t tell the whole story, and how to test for biases that might otherwise go unnoticed. You’ll leave with practical questions to ask when using, building, or evaluating LLM applications.
About Mariia : Mariia Vladimirova is a senior research scientist at Criteo AI Lab and a member of the joint Fairplay team with Inria. She studies trustworthiness in AI systems, including how to spot and measure bias in large language models, and was named a Rising Star in AI Ethics in 2025.

Get ready for the lightning talks:
You have something you want to share with the community or just practice your public speaking? Come on stage and during the 3 minutes time-slot tell us about it.
Some ideas on what you can talk about:

• the project you have been recently working on
• Python library or function you love or which you recently discovered,
• article you've read
• your journey into Python
• conference you have attended

You can decide anytime before the start of lightning talks or you may want to prepare up to one slide (in pdf format) which you can send us the latest on the 28th of September to [paris@pyladies.com](mailto:paris@pyladies.com)

Criteo will be our host and sponsor of the food and the drinks during the networking session after the talks: thank you 💚 and special thanks to Andrey from Criteo for all the support.

Important info

#1:❗For safety reasons, the venue's staff will check everyone's identity on site. 📝Please remember to bring an ID with you and register for the event with your real name and family name. Thank you!

#2: Please be on time. We can’t guarantee a seat once the meetup has started

# 🔍 FAQ

Q. I'm not female, is it ok for me to attend?

A. Yes, PyLadies Paris Tech Talks are open to everyone at all levels.