Software I built for problems I actually had — a project finance diligence agent, résumé tailoring, notes-to-document workflows, quant research infrastructure, and receipt splitting. Some open, some private. The code is the artifact; the README is the argument.
A vertical AI agent for U.S. renewable energy project finance — parses PPAs, validates sponsor models, checks IRA-era tax equity structures, and drafts IC memos with citations. Phase 0: eval harness on hand-graded public PPA cases, not a product yet.
An AI rewrites the bullets in a LaTeX résumé to match a specific job description — without making things up. I built it during the SIPA on-cycle market when every tailoring round took an hour by hand. Pairs with Claude Code (via `claude -p`) or any LiteLLM-compatible model.
A notes-first workspace for turning meetings, messy notes, and audio into structured bilingual documents. It pairs a Markdown editor with transcription, project memory, AI synthesis, and export paths for PDF, Markdown, and Feishu.
A systematic research and execution stack for the A-share market: factor research tracks, backtesting, walk-forward validation, paper trading, and risk controls wired into one research loop.
A receipt-splitting app: upload a receipt, let Qwen VL read the line items, invite friends to claim what they ordered, and settle tax, tip, and shared items without spreadsheet theater.