Ashpreet Bedi: “Investment Committee: 5 Multi-Agent Architectures in Action” / X
Ashpreet Bedi describes building an AI investment committee with seven specialist agents using the Agno framework to manage a $10M US equities portfol...
Md Ismail Šojal 🕷️ on X: “LEANN, an open-source vector database, compresses RAG indexes by an impressive 97% using graph-based recomputation and on-demand embedding calculation. - https://t.co/NhpwzDLDYu” / X
x.comSaved December 30, 2025
Summary
LEANN is an open-source vector database that compresses RAG indexes by 97% through graph-based recomputation and on-demand embedding calculation, eliminating stored embeddings.
It builds on HNSW with two-level graph traversal, dynamic batching for low latency, and graph pruning for minimal metadata storage. LEANN achieves 90% recall in under 2 seconds on large datasets while outperforming baselines like Edge-RAG in storage and speed.
