Hi, I'm Maxime.

I'm a research scientist at Cursive AI, and a PhD candidate in machine learning.

My work has recently involved building long-running software engineering and research agents. This spans LLM post-training, coding eval generation for frontier models, and automated agent harness optimisation.

Some topics I'm interested in include:

• Using probabilistic machine learning to guide decision making under uncertainty, optimising resource use in autonomous agents,
• deep learning theory and understanding training dynamics and architectural choices,
• open-endedness and self-improving AI systems,
• using LLMs to make 'judgemental forecasts' or answer questions with uncertainty; so as to leverage large pre-trained models and unstructured data as a supplement to traditional numerical data in forecasting or decison making