Building enterprise-grade Foundation Model applications

Abstract

Foundation models (FMs), such as Large Language Models (LLMs), have revolutionized software development by enabling new use cases and business models. The unique properties of foundation model applications (e.g., prompts, agents, and the need for orchestration), coupled with the intrinsic limitations of FMs (e.g., hallucination) lead to a completely new set of software engineering challenges. Based on our industrial experience, we identified several key challenges that have caused enterprise foundation model application development to be unproductive, costly, and risky. In this talk, I will discuss these challenges in detail and state the envisioned path for innovation.

Date
Apr 26, 2024 3:00 PM — 3:30 PM
Location
Polytechnique Montreal
2500 Chem. de Polytechnique, Montréal, QC H3T 1J4
Dayi Lin
Dayi Lin
Staff Researcher - Huawei Centre for Software Excellence

Dayi Lin is a Staff Researcher at the Centre for Software Excellence, Huawei Canada, working on the topics of SE for AI and foundation model applications. His research interests include SE for AI, AI for SE, game engineering, and mining software repositories. His work has been published at several top-tier software engineering venues, such as TSE, TOSEM, EMSE, and ICSE. His work on SE and games has also attracted wide media coverage including Kotaku, PC Gamers, Gamasutra, and national newspapers. He received a Ph.D. in Computer Science from Queen’s University, Canada.