Offline search relevance testing with Vespa & Quepid

Hosted by Charlie Hull and Trey Grainger

93 students

In this video

What you'll learn

Why offline relevance testing is a vital search tuning tool

Online A/B testing for search can take weeks to produce results, but what if B really isn’t much of an improvement on A?

How to get Vespa to return matching and scoring information

Vespa is a highly powerful and scalable AI search engine - but it can be complex. We'll show you how to get started.

How to hook up Vespa to Quepid for relevance testing

Quepid is a popular tool that provides a ‘search workbench’. We'll show how to hook it up to Vespa.

Why this topic matters

Vespa is a powerful and scalable open source search engine. Although Vespa provides some offline evaluation tools in Python, there isn’t an easy, interactive way to tune search queries. Quepid provides this for many search engines as a free service based on open source code - but it isn’t obvious how to get the data we need from Vespa’s API. We'll show you how to get started using these tools.

You'll learn from

Charlie Hull

Senior search consultant

Charlie Hull, The Search Juggler, is a search consultant with 25+ years experience of helping companies large and small build powerful, scalable and accurate search applications. His clients include governments, e-commerce giants, recruitment and legal companies across the globe. He is the co-author of “Searching the Enterprise” from NOW Publications, an accomplished conference speaker and organiser and co-hosts the London Search & AI Meetup.

Trey Grainger

Founder @ Searchkernel, Author "AI-Powered Search"

Trey is lead author of the book AI-Powered Search and is the founder of Searchkernel, a software company building the next generation of AI-powered search. He is an advisor to several startups and adjunct professor of computer science at Furman University. He previously served as CTO of Presearch, a decentralized web search engine, and as chief algorithms officer and SVP of engineering at Lucidworks, an search company whose search technology powers hundreds of the world’s leading organizations.


Learn with Trey in the upcoming

AI-Powered Search: Modern Retrieval for Humans & Agents course.

Previously at

OpenSearch Project
OpenSource Connections
Searchkernel
Lucidworks
CareerBuilder

Go deeper with a course

AI-Powered Search: Modern Retrieval for Humans & Agents
Trey Grainger and Doug Turnbull
View syllabus