Abstract

The paradigm of Generative AI is rapidly evolving from passive assistants to autonomous entities. Moving beyond the chatbot and simple code-completion copilots, the next generation of AI agents functions as digital colleagues capable of end-to-end task execution. This talk explores how these agents leverage code not merely as an output, but as their fundamental reasoning engine and interface for interacting with the world. We will focus on the shift from manual prompting to high-level Steering, discussing how human intent can guide code-driven agents to autonomously navigate complex workflows, resolve software engineering challenges, and bridge the gap between abstract requirements and finished, functional results.

Speaker Bio

Qidi Xu is a Researcher at MiniMax, specializing in post-training research for Large Language Models (LLMs) focused on software engineering and terminal usage. As a contributor to the coding capabilities of MiniMax M2 series, he is dedicated to evolving AI from simple code generation tools into “digital colleagues” capable of end-to-end autonomous execution. He leads technical iterations in complex code generation, multi-step task execution, and downstream framework adaptability. By leveraging code as a core reasoning engine, he has significantly enhanced the model’s engineering performance in real-world development environments, enabling it to autonomously navigate complex workflows and bridge the gap between abstract requirements and functional implementation.