The Data Scientist Show - Daliana Liu Making LLMs hallucinate less, how to diagnose ML models, from PM in Google AI to CEO of Galileo - Vikram Chatterji - The Data Scientist Show #066
Aug 1, 2023
Chapters
Transcript
Episode notes
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Introduction
00:00 • 2min
How I Got Into Machine Learning
01:36 • 3min
The Challenges of Putting Large Language Models in Production
04:24 • 2min
How to Diagnose Large Language Models
06:30 • 3min
How to Evaluate LLMs That Aren't Working Very Well
09:57 • 3min
How to Measure Hallucination in Large Language Models
12:46 • 3min
How to Measure Hallucination
15:51 • 3min
How to Diagnose and Detect Hallucination Using Metrics
18:44 • 5min
The Evolution of Open Source Language Models
24:08 • 5min
Galileo for NLP and Computer Vision: A Diagnostics and Explainability Layer
29:33 • 3min
Galileo for Data Scientists
32:44 • 4min
The Risks of Launching a Large Language Model in Production
36:59 • 6min
The Importance of Prompt Engineering in NLP
43:16 • 2min
How to Continue to Learn About the Space
45:03 • 2min
How to Use Galileo to Improve Your Data Science Workflow
46:54 • 5min
Google AI: How to Find Repeatable Use Cases for Machine Learning
51:54 • 2min
How to Build a Custom Machine Learning Solution for Your Customers
53:46 • 5min
How to Refract Your Questions in Data Science
59:15 • 3min
The Importance of Learning From Amazon
01:02:18 • 2min
How to Be an Efficient Product Manager
01:03:53 • 3min
How to Be a Better Manager
01:06:39 • 2min
The Importance of Context Switching
01:09:06 • 3min
The Challenges of Starting a Company
01:12:29 • 2min
How to Start a Company
01:14:27 • 2min
The Importance of Thinking Short and Long Term
01:16:40 • 2min
The Future of Galileo
01:18:42 • 6min
Galileo: The Future of AI Data Science
01:24:24 • 2min