Arun Sekhar

Frontier labs are releasing new models constantly, and it is hard to know when “better” is better enough to justify touching a working system. On top of that, “just swap the model” often turns into real work because providers expose different APIs and different expectations around tools and structured outputs. The model swap workshop is a hands-on bake-off across frontier LLMs. We will run the same scenarios using multiple models (OpenAI, Anthropic, Kimi, and more) and compare results side by side for agentic tool use, structured outputs, and multimodal tasks. Swapping models is not just changing a model name. In this workshop, you will actually do the swaps, including moving between OpenAI-style Responses APIs and Anthropic-style Messages APIs, then see what breaks and what needs to change in your prompts, tool definitions, and JSON strategies. We will finish by running a small eval suite so you can quantify tradeoffs instead of relying on vibes. We will provide the Microsoft Foundry environment for access to the models, no account needed. Speakers: Pamela Fox — Microsoft; Arun Sekhar — Microsoft.

OpenAI SDK, Anthropic SDK, or an LLM-agnostic agent framework. Which one should your next AI app be built on? Starting with Foundry Models, we walk through each option in code, show what you gain and what you give up at every layer, and help you pick the right abstraction for your scenario without overbuilding. Speakers: Arun Sekhar — Microsoft; Pamela Fox — Microsoft.

You already run OpenClaw locally, sandboxed in MXC. Now you need the same agent in the cloud for dev/test, reachable from Teams on your phone, without handing over the keys to the kingdom. This session shows a simple, one‑command path to do all of this: an isolated Container Apps sandbox running an OpenClaw image, calling Azure OpenAI in Foundry Models securely without keys over the standard OpenAI API, scaling to zero when idle.