A 40-query benchmark of AI agent retrieval on Amazon Bedrock: web scraping vs search snippets, compared on cost, latency, and silent coverage loss.
Senior IT leader working on agentic AI, multi-cloud platforms, and AI governance.
I lead cloud infrastructure and generative AI delivery teams at an audio technology company, with 20+ years in software and cloud engineering across AWS and Azure. My writing here is field reports from real builds: multi-agent pipelines on Amazon Bedrock with Strands and MCP, retrieval benchmarking, AI governance frameworks, and cost analysis that goes beyond the whiteboard version. My working principle across all of it: deterministic code computes, LLMs narrate and judge, never the reverse. Away from the keyboard: cycling the trails of northwest Chicago suburbs, Sangam poetry, and Tamil cinema.
Markdown supported · ⌘↵ to post
No responses yet.