Inside Microsoft AI Tour NYC: Proctoring a Hands On GitHub Copilot Migration Lab

On January 21, 2026, I contributed at Microsoft AI Tour NYC 2026 as a workshop proctor at North Javits Center, 429 11th Avenue, New York, NY 10001. This was a hands on role focused on execution: helping builders move from intent to working code, fast, and with quality.

Microsoft AI Tour home
https://aitour.microsoft.com/flow/microsoft/aitour/landing/page/home

NYC event page
https://aitour.microsoft.com/flow/microsoft/nyc26/landingpage/page/cityhome

Workshop repo
https://github.com/microsoft/aitour26-WRK541-real-world-code-migration-with-github-copilot-agent-mode

What Microsoft AI Tour NYC is about

Microsoft AI Tour is built for people who want to do more than just learn concepts. It is designed to help teams translate modern AI and developer tooling into practical outcomes, with a strong focus on real adoption patterns, real constraints, and real delivery.

NYC delivered that energy in person: engineers, architects, leaders, and builders working side by side, sharing patterns, and leaving with concrete next steps.

The workshop I proctored

Workshop: Real World Code Migration with GitHub Copilot Agent Mode

This is a practical workshop that uses a real migration scenario to teach a repeatable workflow for modernization using GitHub Copilot Agent Mode.

The lab focuses on migrating an HTTP API from Python to C# while keeping control over correctness, behavior, and developer experience. Participants learn how to use Copilot in a disciplined way so the assistant accelerates the work without taking ownership away from the engineer.

What participants built

1. A migrated HTTP API, moving from a Python implementation to a C# implementation using .NET Minimal APIs
2. JSON handling in C# using System.Text.Json
3. Swagger and OpenAPI documentation so the API is easy to explore and integrate
4. Tests using MSTest and WebApplicationFactory to validate behavior and protect the migration

What makes this workshop valuable is not the final code alone. It is the migration loop: plan, translate, validate, test, document, and iterate until the result is reliable.

What I did as a proctor

Workshops fail for predictable reasons: slow setup, unclear troubleshooting, and teams drifting into long detours. My job was to protect learning time and keep the room moving toward a finished outcome.

Here is how I contributed during the lab

1. Environment readiness and setup support so people could start building quickly
2. Fast unblocking on Git, VS Code, .NET, and Codespaces issues
3. Coaching on prompting patterns that lead to clean, testable code
4. Reinforcing quality habits: small changes, frequent validation, tests that confirm parity
5. Helping participants interpret failures, not just fix them, so they learn a reusable approach

This is the same mindset I bring to customer engagements: reduce friction, create momentum, and make outcomes repeatable. When a team finishes a workshop with a working solution and a clear workflow, adoption becomes much easier back at the office.

The bigger takeaway: Copilot is powerful, but the workflow is the product

Modernization is not only a translation problem. It is a risk management problem. The fastest path that breaks behavior is not fast. The best migrations are guided by tight feedback loops and clear ownership.

GitHub Copilot Agent Mode becomes most valuable when teams treat it as an accelerator inside a professional loop:

1. Define the target behavior
2. Translate incrementally
3. Validate constantly
4. Lock in correctness with tests
5. Document for the next engineer

That is what the workshop intended to demonstrate, and it is why I was proud to contribute as a proctor with such a great team!

Want to run this workshop with your team

Everything you need is public.

Workshop repo
https://github.com/microsoft/aitour26-WRK541-real-world-code-migration-with-github-copilot-agent-mode

Microsoft AI Tour home
https://aitour.microsoft.com/flow/microsoft/aitour/landing/page/home