
DataTalks.Club Qdrant 2025 Conference Interviews
Nov 28, 2025
Join Andrey Vasnetsov, co-founder and CTO of Qdrant, as he shares the journey of building a developer-focused vector database during the pandemic. Slava Dubrov from HubSpot dives into AI Signals, discussing the power of embeddings in contextual recommendations. Marina Ariamnova of SumUp reveals her innovative LLM assistant that rapidly translates natural language to SQL, transforming analytics workflows. Lastly, Evgeniya Sukhodolskaya explores retrieval research and education at Qdrant, touching on the future of AI in developer tooling.
AI Snips
Chapters
Transcript
Episode notes
Vectors Are Now A Developer Tool
- Andrey explains Qdrant combines database-style querying with vector similarity to serve developer workflows, not end users.
- Vector search shifted from ML specialists to a standard developer tool that anybody can integrate quickly.
Founding Qdrant During Lockdown
- Andrey built an early prototype during COVID lockdowns when he had free time and traction on GitHub followed quickly.
- That prototype evolved into Qdrant after colleagues and a co-founder helped turn it into an MVP and startup.
Contextual Recommendations With Embeddings
- HubSpot's AI Signals uses embeddings and similarity search to provide context for agents and features like look-alike company search.
- They ensure privacy by restricting searches to a user's data and public datasets separately.
