I am a second-year M.S. student working on causal inference for image processing.
I am currently interested in LLM application development, RAG systems, agent workflows, and AI developer tools.
- LLM application development and RAG systems
- Agent workflows and AI coding tools
- FastAPI backend services for AI applications
- Causal inference and causal discovery for image processing
- Representation learning and generative models
A FastAPI-based AI application project that combines Todo CRUD, LLM-powered task extraction, and a local-document RAG learning pipeline.
Built features include natural language task extraction, OpenAI-compatible LLM calls, Markdown document loading, chunking, embedding search, Chroma vector store, RAG answers with sources, and retrieval evaluation.
Python FastAPI LLM Apps RAG ChromaDB SQLite Docker DashScope
A local Windows tool for restoring Codex Desktop conversations that still exist on disk but no longer appear in the app.
It supports source discovery, session import, metadata repair, automatic backups, rollback, and a GUI workflow for normal Windows users.
PowerShell WinForms SQLite JSONL Local-first Developer Tools
A reusable engineering memory skill for coding agents.
It helps coding agents turn solved errors into reusable notes, so similar problems can be recognized and handled more consistently in later tasks.
Agent Workflow AI Coding Developer Tools Knowledge Management
- Building practical LLM applications with FastAPI
- Learning and improving local-document RAG pipelines
- Exploring agent workflows and AI coding tools
- Connecting causal inference research with visual AI systems
- GitHub: Air000000