Dear MuMax3 Community,
My name is Asad, and I’m currently working as a research intern at a UK-based lab conducting research in neuromorphic computing and spintronics. As part of this internship, I’m developing a domain-specific LLM-powered research assistant to support researchers.
One core feature of this assistant is the ability to autonomously run MuMax3 simulations from natural language prompts, with a debugging agent designed to detect and correct errors in case the simulations fail.
To improve the robustness of this system and reduce LLM hallucinations, I am currently building a Retrieval-Augmented Generation (RAG) pipeline for code correction. For this, it would be immensely helpful to access a list or database of common syntax and compilation errors encountered when using MuMax3.
If anyone has compiled such a resource or would be open to sharing logs, examples, or patterns of frequently encountered issues I would be very grateful. Your support would directly contribute to building more intelligent tools for the MuMax user base and early-career researchers alike.
Best regards,
Asadullah Siddique
MEng (Hons) Electrical & Electronic Engineering, University of Manchester