Qdrant

Most vector databases treat retrieval as a single operation. That's the wrong abstraction. Storing embeddings and returning nearest neighbors is a solved problem. The hard problem is what happens next. We solve it through composable vector search, built in Rust. Today, led by AVP, with Bosch Ventures, Unusual Ventures, Spark Capital, and 42CAP, we're announcing our $50M Series B to accelerate it. Learn more about Qdrant’s composable vector search and our latest funding round here: qdrant.tech/blog/series-b-…

Most teams think continual learning is a training problem. Taranjeet Singh from @mem0ai thinks they're solving the wrong thing. At Vector Space Day on June 11, he'll show how giving agents better memory, not better gradients, is what actually makes them improve over time in production. Vector Space Day is a full-day conference for engineers going deep on vector search, retrieval, and agent memory. Join us in San Francisco. Grab your ticket at luma.com/vsd-sf

Most people pay for insurance and barely understand what it covers. The pamphlet is dense, the insurer’s website is complex, and finding an in-network provider with availability takes more energy than most patients have. @SunnyHealthAI is building a healthcare concierge that insurance companies and care providers offer to their members. Members sign in through their payer, the platform already knows their plan and benefits, and a chat experience helps them find in-network providers and book real appointments through an AI voice agent. The retrieval engine underneath is Qdrant. @SunnyHealthAI migrated from Postgres because the schema rigidity slowed every new metadata field they wanted to add, and fuzzy matching couldn’t handle precise patient queries. They needed hybrid search (must-filters expressed alongside semantic ranking), first-class geo re-ranking, and a flexible payload model for deeply nested provider records. That same retrieval layer now powers patient search and the AI voice agent that books appointments on patients’ behalf. Read how they built it: qdrant.tech/blog/case-stud…

Excited to share that Qdrant will be speaking at the @MistralAI AI NOW Summit in Paris 🇫🇷 Chadha Sridi (Developer Advocate at Qdrant) will present: “Semantic Search on Messy Documents with @MistralAI OCR and Qdrant” We’ll explore how semantic search + OCR can turn noisy, unstructured documents into searchable, usable knowledge using @MistralAI and Qdrant. Looking forward to connecting with the AI community in Paris 🙌 Link: ainowsummit.com

TurboQuant Webinar Reminder Better compression without sacrificing recall 👀 Join us on May 26 to see: → TurboQuant vs SQ/BQ → Real benchmark results → Production tradeoffs → Live technical walkthrough 🗓️ May 26 🔗 luma.com/qdrant-turboqu…

Excited to share that Qdrant is joining #bbuzz26 as a Gold Partner! If you’re attending @berlinbuzzwords this year, come meet the team: → Ewa → Benjamin Costes → Chadha → @krotenWanderung We’ll be talking all things: → Vector search → Production retrieval systems → Real-world AI workloads Plus: - Live demos of Qdrant Edge - Lots of amazing merch at the booth Looking forward to meeting the community in Berlin 🙌 🎟️ Tickets: 2026.berlinbuzzwords.de/tickets/

Meet a few of our speakers for Vector Space Day on June 11 at The Midway. They will cover: - Taranjeet Singh from @mem0 Continual Learning Starts with Memory - Paige Bailey from @GoogleDeepMind Literal Skill Issue: Are SKILLS.md Holding Your Agents Back? - @HubSpot Scaling to Billions - @neo4j Free Your Agent's Mind...with Context Graphs - @llama_index The Document Harness: What Your AI Misses in the 90% - @twelve_labs Beyond the Single API Call: Agentic Video Intelligence And more. Join us for a full day of single-track technical content covering search and AI retrieval, agents and memory, and edge and robotics AI. Get your ticket: luma.com/vsd-sf

🚨 Happening today: Qdrant Office Hours Join us with Kevin, our Qdrant Star for: “Threat Intel with Qdrant” → Vector search for CyberOps → Threat modeling + attack surface analysis → Security data correlation at scale Time: 6:00 PM CEST 📍 Qdrant Discord Link: discord.gg/HyzAECxJs?even… #Qdrant #VectorSearch #CyberSecurity #DevSecOp

It’s here . . . We’ve lined up technical talks on vector search, AI memory, context engineering, or retrieval infra from @awscloud, @arizeai, @HubSpot, @neo4j, @Vultr, @twelve_labs, @Adobe, @Qualcomm, @GoogleDeepMind, @cognee_, @mem0ai, @llama_index, @SlackHQ and more. Join 300+ AI builders for a full day on agents and memory in production, retrieval from cloud to edge, and multimodal AI. Don’t miss Vector Space Day on June 11 at The Midway in San Francisco. Early bird tickets are on sale but end soon: luma.com/vsd-sf

Recruiting is a needle-in-a-haystack problem at massive scale: 200M+ profiles, 1B+ data points, and a definition of "the right candidate" that no keyword filter can express. @Perfect_HQ builds an AI recruiting workforce that runs sourcing, screening, outreach, and candidate conversations end to end. But pure semantic search hit a ceiling. A product manager and a product marketer sit next to each other in embedding space. They're not the same job. Acceptance rates plateaued at 30%, the recruiting industry baseline. @Perfect_HQ moved to Qdrant Cloud with hybrid search and multivector representations, structuring each candidate profile as a set of distinct vectors rather than a single embedding. Combined with LLM orchestration, the results: - Match accuracy from 30% to 99.993% in internal benchmarks - 95-100% acceptance rates across largest customer cohorts - Complex multi-criteria searches complete end to end within the latency budget - Full cycle from recruiter intent to live pipeline in under 2 minutes Read the full story: qdrant.tech/blog/case-stud…

Grep works when you know exactly what you’re looking for. Vector search works when you only know the meaning. This project by Jagriti Rai uses Qdrant Edge to build a semantic memory engine for codebases - enabling developers to search repositories using natural language instead of exact keywords. Instead of: → digging through folders → guessing function names → chaining grep searches You can ask: “Where is retry logic implemented?” and retrieve semantically relevant code instantly. Why this is interesting: • Fully local retrieval with Qdrant Edge • No cloud dependency or external APIs • Semantic memory for large codebases • Low-latency vector search running directly on-device Read here: pub.towardsai.net/beyond-grep-a-…

Vector search with Qdrant ≠ just embeddings Real retrieval systems combine: → semantic search → metadata filtering → structured constraints → fast query execution This deep dive explores how deep query filtering in Qdrant works in production pub.towardsai.net/how-to-use-dee… #Qdrant #VectorSearch #RAG

TurboQuant is now in Qdrant If you’re using SQ or BQ today, there may be a better option! → Similar recall to SQ at ~2× compression → Better recall than BQ at the same storage budget Join our live technical session on May 26: • TurboQuant explained • Benchmarks + demos • When to use it in production luma.com/qdrant-turboqu…

CyberOps meets vector search Join our next Qdrant Office Hours with Kevin Figueroa Threat Intel with Qdrant → Similarity search for cybersecurity → Threat modeling + attack surface analysis → Recon + vulnerability data workflows 🗓️ May 21 🕘 9:30 PM IST / 6 PM CEST 🔗 discord.gg/HyzAECxJs?even…

A nanomedicine startup developing treatments for hard-to-treat cancers needed its researchers to search the entire PubMed corpus, not filtered subsets. Sapu indexed all 28 million PubMed abstracts into a single Qdrant collection. Their AI platform now supports research paper authorship, SOP-aware chatbots, and full-corpus literature search across biomedical abstracts. Researchers query a small filtered slice using metadata or run vector search across the full 28 million records at once. Using this tool, they created seven peer-reviewed papers published using the AI tooling, seen daily adoption from the CEO to the lab floor, and inked a new robotics partnership extending the platform into physical lab workflows. Sapu moved from self-hosted Docker to Qdrant Cloud Premium, gaining SOC 2 compliance (critical for licensing the platform to other biotech companies) and the stability to stop pulling engineering attention away from cancer research. Read the full story: qdrant.tech/blog/case-stud…

Qdrant 1.18 is out, featuring TurboQuant; a new quantization method developed by Google Research. Our TurboQuant implementation is an extended version of the algorithm with borrowed RaBitQ ideas. It offers: 1. Similar recall to Scalar Quantization (SQ), using 2x less memory. 2. Better recall than Binary Quantization (BQ), using a similar storage budget. Additionally, we released two other features: 1) Memory Monitoring 2) Adding and Removing Named Vectors Read more about 1.18 in our blog: qdrant.tech/blog/qdrant-1.… And release: github.com/qdrant/qdrant/…

Agent Skills for Qdrant Vector search ≠ just storage anymore → Structured retrieval workflows → Smarter AI agents → Better reliability in production Watch the session: youtu.be/vpbPhiXQ9Ks?si… Read the blog: qdrant.tech/blog/qdrant-sk…

Search shouldn’t stop at reranking. Qdrant 1.17 introduces the first vector index-native relevance feedback approach 👇 → Push relevance into retrieval itself → Smarter vector search → Real production use cases Watch our Vector space talk: youtube.com/live/Z5j1wx1D5… Read the blog: qdrant.tech/blog/qdrant-1.…

Our own Brian O'Grady (Head of Field Research and Solutions Architecture at Qdrant) joined the Stack Overflow Podcast to break down a question that trips up more teams than you'd expect: when do you actually need semantic search, and when is exact-match the right call? They dig into the real differences between traditional Lucene-powered text search and vector databases, where each one wins (logs and security analytics vs. user-facing discovery), and where Qdrant is headed with video embeddings and local-agent contexts. 🎧 stackoverflow.blog/2026/05/05/wha…

Early bird pricing for Vector Space Day 2026 ends soon! If you've been on the fence: this is the moment. After this, tickets go to full price. June 11 in San Francisco. A full day of technical talks on retrieval, agents, and edge AI, with 400+ engineers from across the ecosystem. Lock in your spot at the early bird rate: luma.com/vsd-sf

Qdrant Office Hours on May 21 This month: Threat Intel with Qdrant 👇 → Vector search for CyberOps → Threat modeling + similarity search → Pentest + vulnerability data analysis 🎙️ Featuring Kevin Figueroa 🕘 9:30 PM IST / 6 PM CEST 🔗 discord.gg/6CV3JfTVaW?eve…