Prashanth Rao
4.1K posts
@tech_optimist
AI engineer @lancedb | prev @kuzudb. Blogging @ https://t.co/gLektr01zQ
11 years ago today, Rust 1.0 was released! Nowadays, 1 BILLION crates are downloaded a day, more than all the downloads in the first 49 months since the release of Rust 1.0 in May 2015! TODAY at 7 PM CEST, we'll go live with @orhundev to celebrate and build stuff!
The repo is public ! github.com/lancedb/lerobo…
1/ Hannes Mühleisen announced Quack — @duckdb's new client-server protocol — at Day 1 of @AICouncilConf. Lance works with it out of the box.
The DataFusion community is coming together in San Francisco for the meetup on May 11th at the Chroma office. Yingjun Wu, our CEO, will talk about “From Stream Processing to Data Compaction and Lake Analytics.” Along with other great speakers: Pranav Toggi, Software Engineer at @wherobots: Making Spatial Data a First-Class Citizen in DataFusion Prashanth Rao, AI Engineer at @lancedb: Distributed Query Execution with DataFusion and Lance Stu Hood, Head of Engineering at @paradedb: DataFusion for Top-K Joins You’re building in data infra or databases, this is the Meetup to attend on May 11th in SF. Thanks to Divya Ranganathan, Eddie A. Tajeda, and Andrew Lamb for all their great efforts. Register here: luma.com/k3ointcl
I say this all the time. Databases aren’t sold on golf courses anymore. How you post your measurements and benchmarks matter. These days a lot more choices are made by engineering teams not executives on the back nine after 6 beers. The audience is nerds who care.
Hear ye, hear ye — the LanceDB team rides forth to @AICouncilConf in SF next week ⚔️🏰 Seek us at our booth within the walls of the Marriott Marquis. There, our knights of AI shall demonstrate how the multimodal lakehouse collapses five unruly systems into one sovereign table — so your researchers stop waiting and your GPUs stop starving. We'll be guarding the gates of the multimodal lakehouse. Come find us. aicouncil.com/sf-2026
Super excited about our own coconut @cocoindex_io made to 7k github stars. CocoIndex is the incremental context engine to continuously build search index, knowledge graph and structured insights for your ai agents - codebase, meeting notes, slack, your dynamic corpus for AI to reason and make effective decisions. It's all about building the community and ecosystem around a great open source project. Thanks all the users, the community and everyone for your contribution, suggestions, and love! We are super grateful for our community and are excited to build together with the community down the road!
1/ Multimodal RAG agents typically need 3 hops to reason over media: vector search → graph traversal → blob fetch. @tech_optimist & David Hughes at KGC (Knowledge Graph Conference) propose collapsing all three with Lance, multimodal lakehouse format.
1/ Scaling vector search to 10B vectors has three distinct bottlenecks: linear centroid scans, O(d²) RaBitQ rotation at high dimensions, and index builds that don't parallelize. Here's how LanceDB Enterprise solves each:
