Track1: GenAI Mastery

A Day in the Secret Lives of LLM agents

We would like to delve into the secret lives of LLM Agents a.k.a applications powered by LLMs (Large Language Models). LLMs suffer from a limitation due to the data they have been trained on. One way to circumvent this issue is by connecting the model to external services to provide more meaningful responses.

An agent has access to an LLM and a suite of tools for example Google Search, Python REPL, math calculator, weather APIs, etc. We will also touch upon use cases with the help of Azure Cognitive Services. The workshop will be modeled as a puzzle/case which will require some detective work by the LLM Agents to solve.

Learning goal

The participants will learn how to integrate Agents powered by an LLM model with external services using toolkits. This is not a workshop on understanding the technicalities of LLMs but rather exploring use cases possibilities.

Pre-requisites

Basic knowledge of NLP, Generative AI

  • Abhishek Saxena

    Abhishek holds a PhD in Mechanical Engineering from ETH Zurich. During his studies he received the Excellence Scholarship and the Outstanding Young Academics award from the Swiss Government. Abhishek is an experienced engineering professional with extended experience as a researcher in the fields of Algorithm development, Data Analytics and Mathematical Modelling. Prior to joining D ONE, he worked as a Senior Simulation & Modelling Engineer, in ams Osram AG. Abhishek has been with the team since 2022.

  • Nino Weingart

    Nino holds an M.Sc. and B.Sc. in Computer Science from Swiss Federal Institute of Technology (ETH). Before joining D ONE, Nino worked as a Software and Machine Learning Engineer in Switzerland. Nino also works as a Lecturer on the topic “AI for Manager” and has been with the team since 2021.

  • Nikolaos Mourdoukoutas

    Nikolaos received his M.Sc. in Applied Mathematics from ETH Zurich, with a specialization in artificial intelligence. Before joining D ONE, his industry experience was primarily in biopharmaceutical applications of machine learning, while in academia, he focused on research in Bayesian deep learning. Nikolaos has been with the team since 2022.