Biomedical AI systems fail not because the agents lack capability, but because they lack the right context. Genes, diseases, drugs, and publications don't exist in isolation; they're connected through hundreds of relationships. A search over flat documents retrieves relevant chunks, but can't reason across multi-hop connections. A graph ontology enables that reasoning, but misses semantic relevance. How do you get both? That's the exact problem we're solving with Neo4j on April 21. We'll walk through a live Biomedical GraphRAG AI Copilot, one that combines Qdrant's vector search with Neo4j's knowledge graphs to answer questions like: "Which genes are associated with this disease, and which studies support that connection?" You'll see: - How graph neighbourhoods constrain LLM context (no hallucination, no guesswork) - How hybrid retrieval selects precise evidence from complex corpora like PubMed - How provenance chains link every answer back to source publications Register now: https://luma.com/a7ai1y4t
Qdrant
Software Development
Berlin, Berlin 55,987 followers
Composable high-performance vector search
About us
Powering the next generation of AI applications with advanced and high-performant vector similarity search technology. Qdrant is an open-source vector search engine. It deploys as an API service providing a search for the nearest high-dimensional vectors. With Qdrant, embeddings or neural network encoders can be turned into full-fledged applications for matching, searching, recommending, and much more. Make the most of your Unstructured Data!
- Website
-
https://qdrant.tech
External link for Qdrant
- Industry
- Software Development
- Company size
- 51-200 employees
- Headquarters
- Berlin, Berlin
- Type
- Privately Held
- Founded
- 2021
- Specialties
- Deep Tech, Search Engine, Open-Source, Vector Search, Rust, Vector Search Engine, Vector Similarity, Artificial Intelligence , Machine Learning, and Vector Database
Products
Qdrant
Machine Learning Software
Qdrant develops high-performant vector search technology that allows everyone to use state-of-the-art neural network encoders at the production scale. The main project is the Vector Search Engine. It deploys as an API service, providing a search for high-dimensional vectors. With Qdrant, embeddings or neural network encoders can be turned into full-fledged applications for matching, searching, recommending, and many more solutions to make the most of unstructured data. It is easy to use, deploy, and scale, blazing fast and accurate simultaneously. Qdrant engine is open-source, written in Rust, and is also available as a managed Vector Search as a Service https://cloud.qdrant.io solution or managed on-premise.
Locations
-
Primary
Get directions
Berlin, Berlin 10115, DE
-
Get directions
New York, New York, US
Employees at Qdrant
Updates
-
Qdrant reposted this
GPU is suitable not only for model inference, but also for ANN Indexing. Qdrant OSS supports GPU for faster HNSW Index construction since v1.13. Now clusters with GPUs are also available on cloud.qdrant.io. The difference? Index-building speedup by x5-x10, depending on the chosen hardware setup. For heavy-indexing use cases, it is significant. How to run Qdrant with GPU support ⤵️ https://lnkd.in/dgrX-4zg
-
-
Qdrant reposted this
Slides from my talk at GenAI Zürich on 𝐄𝐝𝐠𝐞 𝐀𝐈 𝐚𝐧𝐝 𝐞𝐧𝐚𝐛𝐥𝐢𝐧𝐠 𝐬𝐞𝐦𝐚𝐧𝐭𝐢𝐜 𝐬𝐞𝐚𝐫𝐜𝐡 + 𝐨𝐧-𝐝𝐞𝐯𝐢𝐜𝐞 𝐦𝐞𝐦𝐨𝐫𝐲👇 We explored when it makes sense to move AI to the edge. With more capable Small Language Models being released, what we can run on-device is expanding quickly. A key piece: 𝐯𝐞𝐜𝐭𝐨𝐫 𝐬𝐞𝐚𝐫𝐜𝐡 𝐚𝐭 𝐭𝐡𝐞 𝐞𝐝𝐠𝐞 → ultra-fast search ⚡ → better context → less hallucinations You can enable this with Qdrant 𝐄𝐝𝐠𝐞 ✨: https://qdrant.tech/edge/ Beyond context retrieval, vector search can power use cases like recommendation systems and anomaly detection. We walked through a concrete example by Thierry Damiba: a video anomaly detection system for surveillance built on a two-tier architecture: → Edge: fast triage with Qdrant Edge on NVIDIA Jetson. → Cloud: deeper analysis with Qdrant Cloud + Twelve Labs (Marengo, Pegasus) on Vultr GPUs. 👉Try it out: https://lnkd.in/e7U8zGk2 👉Build it yourself (great 3-part series by Thierry ✨): https://lnkd.in/eh8cCwSf 👉GitHub repo: https://lnkd.in/ewYTuRsA Curious to hear if you’re exploring Edge AI setups 👀 #Qdrant #GenAI #EdgeAI #VectorSearch #GenAIZurich2026
-
Qdrant Office Hours - April 16 We’re excited to host our next Office Hours session on April 16, featuring a great real-world talk from the community. This time, Tony Skorik will present: “Vigilante - Cluster Guardian for Self-Hosted Qdrant” If you’re running Qdrant in production, this one’s especially relevant. Over the past 2+ years, Tony and his team have been managing multiple self-hosted Qdrant clusters in production. Along the way, they built internal tools to handle scaling, updates, and monitoring - which eventually evolved into Vigilante. 👉 A dedicated management layer for Qdrant clusters 👉 Think Patroni, but for Qdrant What you’ll learn: - Managing and scaling self-hosted Qdrant clusters - Handling updates and maintenance in production - Monitoring cluster health and performance - Lessons from running Qdrant at scale 🗓️ April 16 📍 Qdrant Discord Come join, ask questions, and learn from real production experience 👇 🔗 https://lnkd.in/gsUR2yVR #Qdrant #VectorSearch #OfficeHours #AIInfrastructure #DevOps #RAG
-
-
Qdrant reposted this
Still buzzing from the energy at GenAI Zürich 2026 ✨ I had the chance to attend with Andre Zayarni and Kai Kühnel from the Qdrant team. And what an amazing two days those were! From thought-provoking keynotes to hands-on workshops, inspiring talks, and great conversations, the event really brought together a diverse and high-quality mix of perspectives on GenAI. I also gave a talk on Edge AI and how to enable semantic search and on-device memory with Qdrant, will share more on that (and the slides 👀) in a separate post. One of the highlights of the conference was the hackathon, where I had the fun opportunity to serve on the jury alongside Sabine Wildemann Isabel Steiner, Daniel Naeff, Patricia Ferreiro, Michael Wegmüller, Olivier Cornet, Dino C., Ives Huwiler, Dr. Dr. Christian Hugo Hoffmann, and Ruggiero Dargenio. It was incredible to see the creativity and technical skill on display from all the teams. Well done to all the participants and congratulations to all the winners. A special shoutout to the winner of the Qdrant challenge Ali Aoun for his project Sentinel: Conflict Misinformation Early Warning System👏 Big thanks to the organizers and volunteers for making this such a well-run and engaging event. #GenAIZurich2026 #AI #Hackathon #Qdrant
-
-
We're #hiring a new Solutions Engineer (EMEA) in Germany. Apply today or share this post with your network.
-
Happening Today: Agent Skills for Qdrant We’re going live today with a deep dive into Agent Skills - and how they’re changing the way we build retrieval systems. We’d love to hear your thoughts. Speaker: Our Senior Developer Advocate, Thierry Damiba! We’ll discuss: - What Agent Skills are - How they can improve reliability and consistency - Real-world use cases + demos - Ideas from the community 🗓️ Today 🕤 8:30 PM IST / 5:00 PM CEST / 8:00 AM PDT 📍 Qdrant Discord: https://lnkd.in/g3BJ7jRt Join us, ask questions, and learn directly from the team 👇 See you there 👋 #Qdrant #VectorSearch #AgenticAI #RAG #AIEngineering
-
-
Qdrant reposted this
Close to hitting 20,000 followers/connections on LinkedIn - and I couldn’t be more grateful for this journey. To celebrate, I have something special for all of you! I started posting from my very first year of college. Attended hackathons, went to events, worked with amazing startups and companies - and shared everything along the way. That consistency helped me build strong connections across different domains. I truly believe in quality over quantity, and I’m grateful that this network is filled with genuinely talented and driven people. Your credibility is shaped by the people you surround yourself with - and I’ve been lucky on that front. Now, a small gift for all of my followers 😉 (specially who are in the search space) We at Qdrant are looking for Qdrant Stars ⭐ - This is our ambassador program for people who: - Are building on Qdrant - Actively contribute to the community - Haven’t yet been incentivized for their efforts We want to support you - whether it’s: - Speaking at conferences - Hosting events or hackathons - Creating content - Growing the ecosystem in any way You can learn more here: https://qdrant.tech/stars/ https://lnkd.in/g-KMQJge If you’re someone (or know someone) who fits this, feel free to reach out. My DMs are always open - especially for anything around community, hackathons, or collaborations. At the end of the day, the community has given me a lot - and I’d love to give back ❤️ See ya!
-
-
See You at HumanX 2026 - San Francisco We’re excited to be at HumanX 2026, bringing together 6,500+ leaders, builders, and investors focused on real-world AI. If you’re around, stop by and catch: Thierry Damiba, our Senior Developer Advocate Engineer 📍 Vultr Booth 🕛 ~12:15 PM (PDT) What’s the session about? Thierry will walk through an edge-to-cloud video anomaly detection system built with Qdrant and Vultr GPU infrastructure. The system focuses on detecting what looks different from normal in video streams using: - Vector embeddings + similarity search - Edge-based video processing - Qdrant Edge for on-device retrieval - GPU-accelerated indexing + scalable cloud storage 👉 A practical look at how to move from prototype → production for real-time video intelligence systems. If you’re at HumanX, come say hi 👋 #Qdrant #HumanX #VectorSearch #EdgeAI #AIInfrastructure #ComputerVision #SanFrancisco
-
-
Qdrant reposted this
We're hosting a live session on Tuesday with Thierry Damiba on Qdrant's new Agent Skills! We'd love to hear from you and share thoughts/ideas on these new Skills. Thierry was lead design for these and can share a lot of our learnings. Join us on Discord on Tuesday! https://lnkd.in/ekMwq6C6