Qdrant develops a vector search engine designed for production AI systems, enabling teams to configure retrieval, ranking, and filtering to support scalable applications such as semantic search and AI workflows.
Qdrant, an open-source vector search engine, has closed a $50 million Series B funding round led by AVP, with participation from Bosch Ventures, Unusual Ventures, Spark Capital, and 42CAP.
Vector search initially emerged as a technique for retrieving nearest neighbours from dense embeddings within relatively static datasets. However, modern AI systems operate under more dynamic conditions. Retrieval is now often embedded in agent-based workflows that execute large numbers of queries across multiple data types while interacting with continuously evolving datasets.
Applications such as retrieval-augmented generation (RAG), semantic search, and agent-based reasoning require retrieval systems capable of operating reliably at production scale. Tools designed primarily for single-vector similarity or built on legacy indexing architectures can struggle under these demands.
Qdrant has been developed to address these changing requirements. Built in Rust, the system treats retrieval as a set of modular components (including indexing, scoring, filtering, and ranking) that engineers can configure and combine.
This composable approach enables teams to work with dense and sparse vectors, metadata filters, multi-vector representations, and custom scoring functions while controlling how these elements affect relevance, latency, and cost. By exposing these options, the platform allows search performance to be adjusted to priorities such as accuracy, speed, or efficiency without requiring major architectural changes as workloads evolve.
AndréZayarni, CEO and co-founder of Qdrant, said that many vector databases were originally designed simply to store dense embeddings and retrieve nearest neighbours, capabilities that are now considered a basic requirement:
Production AI systems need a search engine where every aspect of retrieval – how you index, score, filter, and balance latency against precision – is a composable decision.
That’s what we’ve built, and what developers and enterprises are looking for as they scale internal and external AI workloads. This funding accelerates our ability to make it the standard.
The new funding will support the further development and adoption of Qdrant’s composable vector search platform as infrastructure for production AI systems.