News
Updates, product notes and practical coverage for AI platforms and operators.
AWS has published a CDK architecture for agentic retrieval using Managed Amazon Bedrock Knowledge Bases, AgentCore Gateway and Runtime, and the AgenticRetrieveStream API. The design lets an agent choose whether and where to retrieve, revisit sources and return cited answers without operating a vector database. It also adds observability for variable multi-step retrieval paths.
AWS has published a practical blueprint for university assistants grounded in catalogues, syllabi, policies and advising material. Bedrock Knowledge Bases supplies retrieval and citations, while metadata filters, evaluation sets and Guardrails limit topics, prompt attacks and personal-data exposure. Institutions can choose managed or self-managed retrieval according to cost, control and data needs.
AWS has detailed a dual-layer access-control design for retrieval-augmented generation that combines indexed permissions with live checks against authoritative sources. The approach is designed to keep Amazon Quick and Bedrock answers aligned with current SharePoint, Google Drive and Confluence permissions, reducing the risk of stale access rules exposing sensitive material.
Cohere has opened a limited private beta of Compass Cloud, a managed version of its enterprise search and retrieval platform. Developers can use APIs, an MCP server or a Python SDK while Cohere operates the pipeline and model inference. Self-hosted Compass remains available. Buyers should test relevance, access controls, latency and data-location requirements before moving production retrieval.
Pinecone has made full-text search generally available, combining keyword ranking and text filters with vector fields in one index. The release targets retrieval tasks where exact identifiers matter. Current documentation sets out API requirements and important limits, including schema planning, separately supplied embeddings and the distinction between filtering and ranking.
Qdrant 1.19 introduces a four-bit Turbo4 vector datatype, a unified memory configuration and more precise tenant and text filtering. The release can substantially reduce storage, but compressed vectors trade some recall for efficiency. It also removes legacy query endpoints, so self-hosted users should plan staged upgrades and client compatibility checks.