How Reddit Evaluated and Selected Milvus for ANN Vector Search
Learn how Reddit evaluated 11 vector-search solutions and tested finalists with approximately 340 million post vectors. Reddit selected Milvus for its scalability, workload isolation, operational reliability, and fit with its engineering needs.
How Read AI Built Its Enterprise-Ready Agentic AI Search Infrastructure on Milvus
Read AI is using Milvus, an LF AI & Data project, to power enterprise-scale agentic search across meetings, email, chat, CRM, and other collaboration data. In this case study, learn how Read AI built a narrative-aware retrieval layer that supports millions of users and billions of records while delivering sub-20–50ms latency, a 5× improvement in agentic search speed, and the foundation for a more proactive, always-on enterprise AI experience.
Bringing Enterprise Document AI to IBM’s Workforce with Docling
Discover how IBM’s CIO organization integrated Docling into AskIBM to make PDFs, Word documents, PowerPoint presentations, and images fully searchable for more than 280,000 employees. By adding Docling to its document ingestion pipeline, IBM unlocked 250,000 new knowledge passages and demonstrated a scalable, open source approach to enterprise document AI.
Streamlining Neurodivergence Diagnosis with OPEA Project
This case study showcases how the LF AI & Data project OPEA powers a real-time, multi-modal AI system to accelerate diagnosis and personalized treatment for neurodivergent individuals. By orchestrating AI services for language, facial, and vocal analysis, the system delivers clinical-grade performance, maintains strict data privacy, and enables early intervention. Future plans include sensory-sensitivity detection, sign-language support, and edge deployment for faster, more accessible care.
Share Your Story
Are you using an LF AI & Data project in your work? We’d love to collaborate with you to feature your work on this page. Contact us at marketing@lfaidata.foundation to share your story
