For the last 5+ years, I’ve been building web apps that people use: customer-facing interfaces, internal tools, dashboards, collaboration products, and the systems behind them. On my own products, I’ve been the solo engineer from the first idea to production, understanding the problem, shaping the solution, writing the frontend and backend, deploying it, and improving it after launch.

My strongest foundation is product engineering: React and Next.js on the frontend, Node.js and Python on the backend, and the database, infrastructure, testing, and deployment work that makes an application dependable.

AI is part of my recent work, but it is not the whole story. I use it when it makes a product more useful, such as automating repetitive work, searching business data, or turning natural language into a practical workflow.

• Web applications and SaaS products: multi-tenant systems, permissions, billing, dashboards, and real-time collaboration
• Frontend engineering: responsive interfaces, reusable design systems, accessibility, and performance
• Backend engineering: APIs, data models, authentication, integrations, background jobs, and production systems
• Product ownership: taking a product from the first idea through development, deployment, and iteration
• AI features: agents, retrieval, structured extraction, and workflow automation

• Protawk, Solo Engineer, Product & Engineering
  A project management and team workspace for freelancers, agencies, and small teams.
• Tweetpilot, Solo Engineer, Product & Engineering
  An AI copilot that helps founders find the right conversations and write replies in their own voice.
• Trackwise, Solo Engineer, Product & Engineering
  A finance assistant that turns natural-language messages into expense tracking and budgeting actions.
• FasterHQ, Software Engineer
  A business platform bringing websites, customers, marketing, and everyday operations into one workspace.

I also work on AgentDock, reusable TypeScript infrastructure for building agent-powered applications. The project focuses on the engineering underneath agents: agent loops, tool registries, memory, compression, provider helpers, contracts, checkpoints, and reliable execution across application boundaries.

It is a monorepo of independently publishable packages, designed to give developers a clearer foundation for building, testing, and running agent workflows in real products.

AI: OpenAI, LangChain, RAG, agent workflows, evaluation, cost tracking
Product & engineering: SaaS architecture, REST APIs, real-time systems, authentication, billing, testing, CI/CD

I like owning problems end to end. I start by understanding what needs to exist, make the trade-offs explicit, and build in small, testable pieces. I care about the details users feel, including speed, clarity, and reliability, as well as the details developers inherit six months later.

• Good software starts with a clear problem.
• A strong frontend still needs a strong system behind it.
• Product decisions and engineering decisions are connected.
• Shipping is part of the engineering job.

Have a product idea, a web app, an AI workflow, or a codebase that needs untangling?