AI Agents Guides & Articles
An agent needs more than a model call. This collection explores the application around it: memory, tools, user boundaries, prompts, and tests. It also covers coding agents, where the output is a change to software rather than a conversational answer.
Begin by building an agent locally, then examine how instructions and skills affect its behavior. Treat a working demo as a starting point: compare outcomes against a baseline and investigate failures before introducing more tools or guidance.
By Sajeetharan Sinnathurai, Principal Product Manager at Microsoft.
Start here
Build an AI Agent with Azure Cosmos DB: End-to-End Guide
Build a TypeScript AI agent with Azure Cosmos DB, React, durable memory, tenant isolation, diagnostics, tests, and a clear path from local development to Azure.
Smart Prompting, Stronger AI Agent Outcomes — What Actually Works
Over the past few months, one prompting pattern has consistently produced better AI agents for me. Not bigger prompts. Not more skills. Just a better separation of responsibilities.
What Six Months of Building AI Agent Skills Taught Me About Writing for Machines
Lessons from six months building AI coding agent skills: focused domain guidance, realistic evaluation, and knowing when more instructions make outcomes worse.
All AI Agents articles
Newest first. Publication dates distinguish recent guides from historical walkthroughs.
Build an AI Agent with Azure Cosmos DB: End-to-End Guide
Build a TypeScript AI agent with Azure Cosmos DB, React, durable memory, tenant isolation, diagnostics, tests, and a clear path from local development to Azure.
Building DevGlobe with Azure Cosmos DB and AI Agents
How DevGlobe uses Azure Cosmos DB, vector search, MCP, and Azure Functions to power an open-source talent graph for developers and AI agents.
Two Cities, Two Talks, One Agent-First Future: Bengaluru and Chennai
Reflections from two back-to-back community conferences in Bengaluru and Chennai, where I spoke about trusted MCP tools and agent-first development with Azure Cosmos DB.
The Regression Tax: Why Your Agent Skills Might Be Breaking More Than They Fix
Are agent skills always worth using? A new paper reveals that regressions offset 59% of gross gains — and the best skills win by breaking less, not helping more.
Smart Prompting, Stronger AI Agent Outcomes — What Actually Works
Over the past few months, one prompting pattern has consistently produced better AI agents for me. Not bigger prompts. Not more skills. Just a better separation of responsibilities.
What Six Months of Building AI Agent Skills Taught Me About Writing for Machines
Lessons from six months building AI coding agent skills: focused domain guidance, realistic evaluation, and knowing when more instructions make outcomes worse.
Joining Scott Hanselman on the Azure Developers Show: Turning Coding Agents into a Cosmos DB Expert
I recently had the incredible opportunity to join Scott Hanselman on the Azure Developers Show to talk about something I've been deeply passionate about —...
Search all blog posts | Complete archive | Subscribe via RSS