The many shades of text search in vector retrieval

Lightning Lessons

Practical skills & tools to accelerate your career

The many shades of text search in vector retrieval

Hosted by Evgeniya Sukhodolskaya and Doug Turnbull

Thu, Oct 2, 2025

4:00 PM UTC (45 minutes)

Virtual (Zoom)

Free to join

By continuing, you agree to Maven's Terms and Privacy Policy.

Invite your network

Go deeper with a course

Cheat at Search with LLMs

Doug Turnbull

View syllabus

Thu, Oct 2, 2025

4:00 PM UTC (45 minutes)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

Cheat at Search with LLMs

Doug Turnbull

View syllabus

What you'll learn

Mismatched Expectations from traditional to vector search

How expectations from a traditional, keyword based search engine don't always translate

Dense vs Sparse Retrieval

What are the differences between dense and sparse neural retrieval with text?

How QDrant thinks about text retrieval

QDrant's core "opinions" about text retrieval (text, vector, filters), and that leads to their implementation

Why this topic matters

Approaching text in vector search is not easy! Text search comes with expectations from traditional search engines that when carried into vector search, lead to unexpected results and dissatisfaction. We’ll discuss the different types of text search from filtering, rule, and BM25-based ranking, to sparse neural and dense vector retrieval how/why each approach is (or isn’t) included in QDrant.

You'll learn from

Evgeniya Sukhodolskaya

DevRel QDrant

A Developer Advocate at Qdrant with four years in Developer Advocacy and eight years in IT. Holds bachelor's & master's in machine learning. Mainly interested in NLP and R&D tasks, she "always ends up working with or in search", from search relevance evaluation with crowdsourcing to N3 RDF query optimisation. Currently a happy certified yapper on vector search, with a focus on search relevance & sparse neural retrieval.

Doug Turnbull

Search consultant

Doug Turnbull guides companies with search relevance and vector retrieval implementations. Doug worked as Principal ML Engineer at Daydream, where he built hybrid search for fashion. Doug led ranking model redesign at Reddit, significantly improving search outcomes. Doug led search at Shopify and served as CTO at OpenSource Connections.

Doug's books include Relevant Search (Manning, 2016) and AI Powered Search (Manning 2024). He created popular open-source tools, including Quepid and the Elasticsearch Learning to Rank plugin.

Find Doug at softwaredoug.com

Previously at