Work

Projects

Selection · code and systems

9: Projects
6: In production

Project · 01 2025–2026 open source

Sterna

A multi-model AI workspace, now open source under AGPL-3.0: side-by-side model chat, a sandboxed coding agent with a GitHub issue-to-PR flow, a RAG knowledge base, MCP connectors, voice rooms, and per-message cost accounting.

Open-sourced under AGPL-3.0: the full Django/DRF backend, React 19 + TypeScript frontend, FastAPI microservices, and the Kubernetes/Terraform infrastructure
Sandboxed coding agent that turns a GitHub issue into a reviewed implementation plan and a pull request, with an in-browser Monaco IDE over its workspace
Every message carries measured tokens, cost, and latency — pgvector RAG, MCP connectors, live multi-agent voice rooms, and BYOK provider keys round out the workspace

PythonDjangoReactOpenRouterpgvectorKubernetesCloudflare

1,264 automated tests

AGPL-3.0 open source

Project · 02 active

Claude Plan Execute: Autonomous Task Orchestrator

A plan→review→implement orchestrator that drives the Claude Code CLI through a multi-phase, multi-agent workflow over a declarative task slate, with quality gates, branch-per-task, cross-task memory, and a subscription-pool-preserving tmux backend.

PythonClaude CodetmuxYAMLpytestFastAPI

Project · 03 active

shallnot: Spec-to-Test Traceability Gate for Coding Agents

A deterministic gate that stops a coding agent from reporting work as done while a requirement still has no passing test, joining specs, test sources and JUnit XML results in a single static Go binary that never calls a model.

GoYAMLJUnit XMLGitHub ActionsClaude Code

Project · 04 in production

CCA-F Exam Trainer: Bilingual Certification Practice

A fully client-side, bilingual (English and French) practice trainer for Anthropic's Claude Certified Architect Foundations (CCA-F) exam: timed mocks weighted across the five official domains, a scaled score with a per-domain breakdown and a full answer review, plus a study mode of original course summaries grounded in first-party Anthropic documentation.

React 19TypeScriptViteTailwind CSS v4ZustandPlaywrightGitHub Actions

Project · 05 active (paused)

Autonomous Research Engine

A self-improving, multi-agent LLM orchestration system that autonomously attacks open problems in mathematics and science, verifying claims formally and producing publication-ready research papers.

PythonClaude CodeLaTeXaiosqliteSymPyZ3FastAPI

Project · 06 MVP ready

Arabic Scholarship RAG Platform

A multi-user platform for citation-exact answers over classical Arabic books, powered by a hybrid BM25 + pgvector retrieval pipeline with cross-encoder reranking, verifier-loop recursion, and a precision "I don't know" threshold gate.

PythonFastAPIpgvectorOpenRouterReactPostgreSQLDocker

Project · 07 active

Atelier: Claude Code Plugin Marketplace

A personal Claude Code plugin marketplace shipping opinionated, reusable workflow skills (from project bootstrapping to interactive-Claude migration) as installable plugins with a consistent structure and resume-aware execution.

JSONMarkdownClaude CodePythonTypeScript

Project · 08 shipped

MCP Secrets Vault: AI-Safe Secret Management

A published npm MCP server that lets AI assistants use secrets (API keys, tokens, credentials) to perform authorized actions without ever exposing the secret values.

TypeScriptMCP SDKZodVitestGitHub Actions

Project · 09 active

Hevy Periodizer

An open-source training-periodization engine you configure instead of code. Goals, equipment and constraints go in; a prescribed week comes out, every load resolved to plates the athlete actually owns. Claude runs the coaching conversation, the engine does the arithmetic.

PythonPydanticHypothesispytestClaude CodeHevy API