11th DOT NET DAY
60 Minutes to Build Your First AI Agent
A 60-minute live build: from a blank file to a working agent loop on Claude — the architecture, the tools, and the exit conditions every team gets wrong.
Thanks & recap
Big thanks to the MDCA team for putting together another fantastic edition of DOT NET DAY, and to the Ahmedabad Management Association for the venue. Speaking to a room of working .NET developers about agentic AI was exactly the kind of conversation I want more of — practical, opinionated, and grounded in code that actually runs.
In this session we built a working agent end-to-end in 60 minutes: the ReAct loop (Think → Act → Observe), tool design, exit conditions, and the part most tutorials skip — what to do when the agent loses the plot. The full presentation and the demo source code are linked below; clone, run, and break things.
If you have follow-up questions, want to bring this content into your team as a workshop, or just want to compare notes on multi-agent architectures — DM me on LinkedIn or WhatsApp.
Jigar Joshi
Jigar is an agentic AI consultant, trainer, and architect who helps teams ship production-grade systems. He trains development teams and consults with IT companies to build autonomous AI agents, and every engagement is customised to the client’s tech stack and produces working deliverables.
He runs Wan Buffer, an agentic AI consulting and training firm, delivering corporate training to development teams, tech leads, and IT company leaders — from keynote stages to hands-on workshops for software CEOs.
Moments
Resources from the talk
Want this format inside your company?
Explore corporate training →



