The MAD Podcast with Matt Turck Making AI Work: Fine-Tuning, Inference, Memory | Sharon Zhou, CEO, Lamini
46 snips
Jul 25, 2024 Sharon Zhou, CEO of Lamini, discusses the evolving AI market, Inference, Memory Tuning, AI agents, real-time challenges, and Lamini's future. Topics include GPU shortage, AMD vs Nvidia, LoRA, and reasoning vs pure compute in AI.
Chapters
Transcript
Episode notes
1 2 3 4 5 6 7
Intro
00:00 • 2min
Evolution of Enterprise AI Applications and Infrastructure
02:03 • 18min
Understanding AI Model Training and Tuning Process
20:25 • 2min
Discussion on LEMONI's Output, GPU Optimization, and Inference Differentiation
22:09 • 3min
Memory Tuning for Improved AI Accuracy
25:32 • 10min
Agents in AI: Leveraging LLM and Case Study on SQL Agent
35:59 • 6min
Debating the Role of Reasoning and Symbolic Components in AI Models
41:52 • 2min