I'm an undergraduate student learning artificial intelligence and trying to understand how intelligent systems are built from the inside.
My current interests are centered around multi-agent systems, LLM agents, and model-level research. I care about not only using AI systems, but also understanding their mechanisms, limitations, and failure modes.
- Machine learning and deep learning fundamentals
- Attention, Transformer, and large language models
- Agent planning, memory, tool use, and evaluation
- Multi-agent communication and collaboration
- Research code reproduction and environment management
This GitHub profile is mainly used to record my learning and research process:
- course assignments;
- reading notes;
- paper summaries;
- small experiments;
- research plans;
- reproducible workflows.
I prefer writing notes in Markdown, because it keeps my thinking process visible and version-controlled.
I am currently exploring multi-agent systems, especially questions like:
- When are multiple agents better than a single agent?
- How should agents communicate?
- How can we evaluate each agent's contribution?
- How do multi-agent systems fail?
- How can agent workflows be made more reliable and reproducible?
-
summer-research-materials
Summer research notes and paper-reading materials on multi-agent systems. -
SJTU-CS2501H-Discrete-Math-Notes
Personal notes for discrete mathematics. -
CS61A-Assignments
Programming practice from CS61A.
- Python
- PyTorch
- Git / GitHub
- Linux / WSL
- Conda
- Markdown
- GitHub: @Aletheiar