Rishabh Nimje·Oct 27Don’t Secure the Prompt.The security model for AI changes dramatically when an assistant becomes an agent.A response icon2A response icon2
Rishabh Nimje·Oct 26Speculate Computation, Not Consequences: Speculative Execution for AI AgentsImagine a coding agent working on a large repository.A response icon1A response icon1
Rishabh Nimje·Oct 24The Most Important Failure Span May Be Green: Observability for Autonomous AI AgentsYour production agent returns the wrong answer.
Rishabh Nimje·Oct 22Your Agent Cluster Is Not a GPU Farm: The Hidden CPU Bottleneck in Agentic AIYou deploy an expensive GPU cluster for autonomous coding agents.
Shivam Ramani·Oct 13I’ve Been Thinking About What AI Should and Shouldn’t Do in UI/UX DesignI think most designers are familiar with the blank-page problem. You open Figma, create a frame, and then spend time figuring out where to…A response icon1A response icon1
Rishabh Nimje·Oct 9Your AI Agent Shouldn’t Have Every Tool All the Time: The Case for Pluggable Agent KernelsImagine an enterprise coding agent whose runtime supports:
Oleksandr Tranchenko·Aug 30Engineering Reliability into AI Agent Code GenerationPart II — The Components
InBeyond LocalhostbyLets Learn Now·Aug 19AI Can Generate Features Overnight — But These Patterns Decide Whether Your System SurvivesThe New Engineering Skill Nobody Expected in the AI EraA response icon1A response icon1
InTowards AIbySandeep Chaudhary·Aug 13From Factory Pattern to ReAct: How Design Patterns Evolved into AI ArchitectureDesign Patterns Never Died. They Moved Up the Stack.
