Pinecone Nexus is in Public Preview.
A knowledge engine that compiles untapped, distributed enterprise data into a layer agents query directly. Answers get faster, cheaper, and more accurate.
Read more: pinecone.io/blog/pinecone-…
A session groups rows by index shape, one EC2 worker per shape. First row in a shape pays the import cost. Every row after reuses the index and reconfigures capacity.
That's what makes a 48-cell sweep run in the background instead of taking a week.
pinecone.io/blog/behind-th…
Same query, two index layouts. Left: old layout, every block lights up. Right: new layout, only the query's terms get read.
Full writeup: pinecone.io/blog/sparse-v3/
Decoupling embedding from ingestion means whatever hardware is on hand can do the work.
One script, runtime device check: MPS on Apple Silicon, CUDA on NVIDIA, CPU fallback otherwise.
10K-20K records per Parquet partition, split across machines, merged, bulk imported.
🔗
We're hosting the August LA Agentic AI Meetup on Tuesday, August 11, 5 to 7pm at Gulp in Playa Vista.
May drew 70 people. July brought 80. Let's break 100 in August.
Come hang out with builders, engineers, founders, and the AI-curious for networking, a few live demos, drinks,