Track1: GenAI Mastery
Red Teaming for Large Language Models
As Large Language Models (LLMs) and Generative AI (GenAI) technologies become integral to our digital landscape, they bring unprecedented capabilities alongside significant risks. The LLM Red Teaming Workshop offers an immersive experience to equip participants with the knowledge and skills needed to anticipate and counteract potential threats in this evolving domain.
By adopting Red Teaming—a concept rooted in cybersecurity—this workshop delves into the vulnerabilities of LLMs and the implications of adversarial attacks. Participants will gain insights into various risk scenarios and engage in hands-on exercises, simulating attack strategies to safeguard LLM systems effectively.
Learning goal
After the workshop, participants will understand various LLM vulnerabilities, ways in which they can be exploited and how they can be safeguarded.
Pre-requisites
Basic Python knowledge, experience working with LLMs would be beneficial but not mandatory. Need to bring a computer.
-
Ali Ander
Ali graduated with a MSc in Electrical Engineering and Information Technology from ETH Zurich. Before joining D ONE he worked as a Data Scientist specialising in machine learning for smart wearables at an ETH spin-off. Ali has been with the team since 2024.
-
Stepan Gaponiuk
Stepan holds a Specialist Degree in Mathematical Methods in Economics from Saint Petersburg State University of Aerospace and Instrumentation. He has been working as a business analyst, BI and ETL Developer and Scrum Master in the past. Before joining D ONE, he worked as a BI Developer at Nordea and a consultant at Accenture. Stepan has been with the team since 2022.