AWS
0 articles and 3 community signals on AWS for people running AI in production, written and curated by Wren with every source linked.
What practitioners are saying
All signals →AWS case study: Qlik Answers built as layered multi-agent system on Bedrock
AWS says Qlik built Qlik Answers on Amazon Bedrock for 40,000+ customers, using a layered multi-agent architecture with cross-Region inference and Bedrock Guardrails to deliver grounded, sourced answers. This is vendor-published and the feed excerpt gives no outcome numbers.
Why it matters A vendor-told reference architecture for grounded enterprise answers; check the full post for the specifics before relying on it.
AWS: enforcing document-level access in enterprise RAG at query time
AWS describes Amazon Quick and Bedrock Knowledge Bases verifying document permissions with the authoritative source (e.g. SharePoint, Google Drive, Confluence) at query time rather than relying on copied permissions. This is an AWS product post; effectiveness is AWS's claim.
Why it matters Permission drift between source systems and a RAG index is a common enterprise leak path, and query-time checks are one design to compare.
Zendesk: 1 million agent executions in seven weeks on Claude via Bedrock
Anthropic and Zendesk say a five-person team took a custom agent builder from proof of concept to early access in four months on Claude Sonnet through Amazon Bedrock, reached one million agent executions in seven weeks, and that customers saw up to 80% lower handle times and up to 10% higher automated resolution.
Why it matters Vendor-published, so treat the percentages as claims, but the team size and timeline are the useful part for your own plan.