AI for the working mathematician, part 5 | Daniel Tubbenhauer

Описание: 🎓 AI for the working mathematician | Daniel Tubbenhauer

What is everyday AI for math? Why should mathematicians care about large language models, generative tools, and “assistants” that hallucinate as much as they help? And what does any of this have to do with the actual daily grind of reading papers, writing notes, running computations, and polishing talks?
In this short series, we explore how AI in 2025 can be folded into real research workflows. The focus is not futuristic speculation, but practical, small-scale uses: triaging arXiv papers, coding quick experiments, sketching proof outlines, drafting LaTeX/TikZ, and building lightweight reproducible pipelines. AI won’t prove fancy theorems for you, but it can be a surprisingly useful (and sometimes dangerously wrong!) collaborator.

💡 Keywords: math research workflow, coding, proof guidance, Sage/Python, LaTeX/TikZ, reproducibility, LLMs in practice
💬 Comments welcome! Corrections, caveats, and stories from your own experiments are very welcome.
Contents of the series will loosely follow:
• What is proving with AI? (outlines, counterexamples, and failures)
• What is coding with AI? (turning definitions into small datasets)
• What is reading with AI? (triaging papers & definitions)
• What is writing with AI? (abstracts, BibTeX, TikZ polish)
• What is reproducibility with AI? (templates, notebooks, project hygiene)

About me.
Hi, I’m Daniel Tubbenhauer (but feel free to call me Dani, they/them). I’m a mathematician working at the interface of algebra, topology, and category theory. My work often involves diagrammatic and categorical approaches to representation theory, quantum invariants, and low-dimensional topology. I’m passionate about visual and conceptual ways of understanding abstract mathematics, and I enjoy sharing that perspective through informal talks, lecture notes, and videos like these.

🌐 Website: http://www.dtubbenhauer.com
📁 TeX and slides: https://github.com/dtubbenhauer/My-Te...
🧵 #MathAI #MathResearch #PureMath