BDBhuvesh Dhimaninblog.bhuveshdhiman.com·Jul 19 · 1 min readSearch built for agents doesn't look like search built for humansIn July 2026, Amazon Bedrock shipped a managed knowledge base built specifically for agentic search, not human search. A human search box expects one query and one ranked list back. An agent needs som00
BDBhuvesh Dhimaninblog.bhuveshdhiman.com·Jul 19 · 1 min readLangSmith just turned every coding agent into one shared traceThat sounds like a minor observability tweak. It is not. Coding agents have been black boxes by default. Each one keeps its own log format, its own session model, its own idea of what counts as worth 00
BDBhuvesh Dhimaninblog.bhuveshdhiman.com·Jul 19 · 1 min readMost coding agents run setup instructions without asking a single questionA published exploit chain shows exactly why that habit is dangerous. A coding agent pointed at a malicious README runs pip install verbatim, against an attacker-controlled package index. The exploit e00
BDBhuvesh Dhimaninblog.bhuveshdhiman.com·Jul 4 · 2 min readTwo labs shipped agent memory frameworks in June 2026. They disagree on one thingMicrosoft and LangChain both shipped agent memory frameworks in June 2026. Both aim at the same wall. An agent that has to be re-fed its whole history gets slower and dumber as the task grows. Both cu00
BDBhuvesh Dhimaninblog.bhuveshdhiman.com·Jul 4 · 2 min readMost teams sandbox agent code in containers. LangChain argues you do not need toMost teams run agent-generated code inside a container and call it sandboxed. LangChain made a different argument on June 30, 2026. You can run untrusted agent code safely without a container at all. 00
