Why We Embed Documents, and Why You Should Go Multi-Vector

Hosted by Ben Clavié and Hamel Husain

Tue, Jun 23, 2026

10:30 PM UTC (1 hour)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

Save 25% til Sunday
AI Evals For Engineers & PMs
Hamel Husain and Shreya Shankar
View syllabus

What you'll learn

Why do we even need retrieval?

Let's take a second to think about the nature of information retrieval: how it evolved, how it needs to continue to evolve, and how tools are synergistic with their users.

What makes single-vector embeddings great, until they aren't

Hear exactly why single-vector methods are omnipresent despite their obvious flaws, and how these flaws limit their usefulness in extremely important, real-world settings.

Where retrieval still breaks, and what's next

Ben's honest take on what late interaction still can't do, the gap to perfect retrieval, and how agents are changing what we ask retrieval to do.

Why this topic matters

Almost everyone reaches for single-vector embeddings, then hits a wall: they dilute long documents and fall apart out-of-domain. Retrieval is about picking the right tool for the moment; single-vector methods no longer fit. Ben Clavié built some of the tools that made multi-vector retrieval usable. He'll zoom out on why he reaches for these models, and the measurable impact they bring to agents.

You'll learn from

Ben Clavié

Member of Technical Staff at Mixedbread

Ben Clavié is a member of technical staff at Mixedbread conducting research on information retrieval, having recently been a major contributor to both the Wholembedv3 model and novel listwise reranking approaches. Previously at Answer.AI, he co-led the ModernBERT project that modernized the widely used BERT backbone and helped democratise multi-vector retrieval methods via the RAGatouille library and models such as answer-ai ColBERT. Based in Tokyo and academically affiliated with the National Institute of Informatics, he also co-organises the Late Interaction Workshop series to further grow the community of researchers exploring methods beyond single-vector embeddings.

Hamel Husain

ML Engineer with 25+ years of experience

Hamel Husain is a ML Engineer with over 20 years of experience. He has worked with innovative companies such as Airbnb and GitHub, which included early LLM research used by OpenAI, for code understanding. He has also led and contributed to numerous popular open-source machine-learning tools. Hamel is currently an independent consultant helping companies build AI products.

See all products from Hamel Husain & Shreya Shankar

Sign up to join this lesson

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