Utilizing LLM Agents for Efficient Requirement Analysis and Specification
By Brain Aboze Elizabeth Ogunyemi
Data ScientistAbstract:
In this tutorial, we will build a practical application using Streamlit, leveraging open-source LLMs, embedding models, and vector stores—all implemented in Python. Attendees will learn how to integrate these components to create a user-friendly interface that enables seamless interaction with LLM agents. The session will cover the iterative development process, highlight the role of human evaluators in assessing output quality, and demonstrate the creation of efficient query engines for summarizing and semantically searching meeting notes. Additionally, attendees will gain a deep understanding of LLM agents, exploring various tools, tasks, and prompts accessible to the agents to enhance their functionality and effectiveness.
GO BACK
Other Talks
-
-
Building Modern AI for African Languages
by Isheanesu Misi -
How Many Wings does the Giraffe Have? - Hallucinations in MLLMs
by Johannes Kolbe -
Pixels to Predictions: Making Math Fun with Python for Real-World Applications
by Paulina Boadiwaa Mensah -
Keynote
by David Mertz