Tracing the journey from GitHub Copilot's early days as a simple autocompletion tool to the current ability to assign complex tasks like test migrations to an AI agent.
The integration of chat and voice allows anyone to code using their native language, breaking down barriers and expanding the developer community.
Developers can now choose from a spectrum of AI tools, from simple autocompletion to powerful synchronous and asynchronous agents that handle entire tasks.
An interactive session demonstrates using an agent in the IDE to add a new snake game to an existing project, including debugging errors through conversation.
A review of a pull request generated by a coding agent shows how it autonomously handled a test migration task and provided a detailed report of its work.
The future of software development involves hybrid organizations where developers orchestrate multiple interacting agents, with AI expected to write 90% of the code.
AI will not replace developers but will act as a superpower, automating tedious work and freeing up developers to focus on creativity, problem-solving, and product vision.
The ultimate goal of AI agents is to foster developer happiness by making software creation more accessible and joyful, enabling a future with a billion developers.
30:24Markus Walker
58:06Anna Fritsch-Weninger
31:39Rainer Stropek
28:40Dr. Alexander Wachtel & Julia Kordick
24:22Philipp Schmid
26:52Thomas Dohmke & Demetris Cheatham
28:09Jan Oberhauser, Mirko Novakovic, Alex Laubscher & Keno Dreßel
29:03Jörg Neumann
Jobs that call for the skills explored in this talk.
autonomous-teaming
München, Germany
Avanade
Berlin, Germany
Avanade
München, Germany
Avanade
Kronberg im Taunus, Germany
