I build.I ship.I bury thedead.

An Electron command deck for developers juggling a pile of Claude Code, Codex and Antigravity projects. It scans git repos under configured folders and surfaces git state, how long each has been neglected, and per-agent session history as cards, resuming your last session in one click. On Windows a Cockpit replaces the wall of external terminals with in-app ones: each session shows its model, active working time, a context percentage that colors up as it nears the limit, and a one-line summary of what it is actually working on — grouped by urgency, so you can see who is waiting on you. When AI refinement is on, each summary is condensed by that session's own agent CLI, so one vendor's quota is never spent on another's work. I treated Electron security as a first-class concern (contextIsolation, sandbox, argv-not-string-interpolation against command injection) and backed it with 465 tests and a Playwright + axe e2e gate.

A five-modality personal generation factory where Claude drives image, video, 3D, voice, and music through MCP tools — generate → machine gate → inspect the thumbnail directly → retry. I consolidated a shared kernel (GPU arbitration, a SQLite asset store, thumbnails, a common server skeleton), per-modality MCP servers, and a FastAPI + React gallery into a single monorepo via history-preserving subtree merges, centralizing the generation GPU, data, and frontend on one RTX 4080 laptop. The gallery now carries an agent harness: describe what you want in natural language and a local LLM calls the tools itself, walking generation through review. Character-preserving edits, a multi-hop lineage tree, collection filters and tag autocomplete make the library browsable. Asset lineage and review verdicts live in the DB so an agent's judgment flows back into the library, locked by 320+ tests. Development runs across two machines, with machine-specific values isolated into profiles and a six-layer fingerprint (code, Python, ComfyUI, models, runtime, machine) answering whether they are actually in the same state. It absorbs and succeeds 3d-asset-studio, which supplied the assets for Studio Apartment.

An open-source, bilingual (EN/ES) web app that estimates which rent law protects an LA renter — City RSO, AB 1482, LA County RSTPO, or just-cause — from their address, plus the current legal increase cap and a rent-increase legality checker. With no public RSO-registry API to lean on, I designed an 'honest estimator' over live LA County Assessor and US Census open data (cross-verified against the county roll), with a four-regime rules engine that leans protective at every ambiguous boundary and degrades to 'pending — confirm with LAHD' rather than ever being confidently wrong. Locked by 222 tests; a profiled indexed-query fallback cut lookups from 13–55s to ~1s. PageSpeed 100/100/100/100, WCAG AA in light and dark.

A RAG system that verifies every citation against the retrieved chunks before it ever reaches the screen, and answers the question of retrieval quality by measuring it rather than asserting it. The corpus is ~3,000 pages of the PostgreSQL 17 manual — full of passages that look nearly identical (pg_dump vs pg_dumpall, five different work_mem settings), exactly where naive vector search quietly returns the wrong paragraph. Across 46 evaluation questions, cross-encoder reranking scores recall@5 0.978 and MRR@10 0.834 against 0.652 for plain vector search, with citation validity at 1.000 and zero hallucinated citations out of 168. Embeddings and reranking run locally through ONNX, so ingestion makes no outbound calls at all.

An open-source Windows tray app that turns PC power use into electricity cost. It combines Windows 11 EMI and PawnIO-backed CPU sensing, NVML/ADL GPU readings, and ACPI battery measurements, while modeling the power and PSU losses sensors cannot see. Flat, tiered, and time-of-use tariffs plus taxes and fixed charges are validated from structured JSON and combined with local SQLite history.