Session
Autonomous AI Agents: From Memory to Self-Modification in .NET
Artificial Intelligence is all the rage, and 2025 is said to be the year of agents. Agents being small AI systems, empowered to act on the users behalf.
This talk provides a practical guide to building AI agents using the .NET ecosystem, and services in Azure like AI Foundry. We're diving in to how to give the agents reasoning, what are the best models, and why? We'll look at the best ways to give agents memory, and the best ways to introduce tool use. How do we give our agents a sense of time, and can we enable them to evolve and improve over time?
This talk is ideal for .NET developers, AI enthusiasts, and anyone interested in the future of intelligent automation. Join me to learn how to build intelligent agents!
All the examples on how to build Agents on Microsoft Azure AI Foundry is in either JS or Python. This talk focuses on the .Net developers fluent in c# that are familiar with Azure.
There will be a fair amount of concepts introduced in this talk, like reasoning, memory, time, tool calling and self improvement. This means that most of the system will be assembled before hand, code will be shown together with services in Azure and thier relations to each other, but in the interest of time, not much live coding will be done.

Andreas Wänqvist
AI Architect, Voyado | Microsoft MVP | Aurelia Core Team
Linköping, Sweden
Links
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top