Vector Podcast

Doug Turnbull - Staff Relevance Engineer, Shopify - Search as a constant experimentation cycle

12 snips
Oct 1, 2022
Ask episode
Chapters
Transcript
Episode notes
1
Introduction
00:00 • 3min
2
The Origins of Cupid
02:56 • 4min
3
How Cupid Improved Relevance in Search
06:36 • 3min
4
How to Create a Successful QA Process
09:09 • 2min
5
How to Structure Your Work at Shopify
10:57 • 4min
6
The Future of Search
14:37 • 6min
7
How to Coach Stakeholders to Understand What You Even Do
20:09 • 4min
8
The Importance of Analysis in Listening
23:52 • 1min
9
How to Balance Precision Versus Recall in Query and Content Understanding
25:22 • 4min
10
The Importance of Tokenization in Search Engines
29:36 • 2min
11
Vector Search: A Moment of Disappointment
31:35 • 3min
12
The Importance of Attention Mechanisms in Search
34:35 • 4min
13
BERT's Inverted Index Bar Search Method
38:05 • 2min
14
The Future of Dense Search
40:31 • 2min
15
The Advantages of a Dense Vector Index
42:15 • 5min
16
The Importance of Filters in E-Commerce
47:05 • 5min
17
The Future of Neural Search
52:10 • 4min
18
The Differences Between Vector and Neural Search
56:34 • 2min
19
How to Engineer a Search Engine With Dense Retrieval Signals
58:19 • 5min
20
How to Be a Successful Data Scientist
01:03:02 • 5min
21
How to Use Gaussian Models to Expand Your Click Base
01:08:27 • 2min
22
How to Improve Your Search Results
01:10:47 • 4min
23
Google's 300 Billion Dollar Opportunity for Everyone
01:14:31 • 2min
24
How to Form a Search Team
01:16:14 • 4min
25
The Role of a Product Manager on a Search Team
01:20:34 • 3min
26
How to Be a Product Manager
01:23:11 • 2min
27
How to Be a Successful Product Manager
01:24:50 • 2min
28
The Future of Data and Engineering
01:26:38 • 3min
29
The Importance of Relevance in Machine Learning
01:29:09 • 4min