Handwork
(2026)
Handwork is an AI coding agent for the terminal and desktop.
You ask it to fix a bug or build a feature. It reads the project, edits files, and runs tests with the model you choose.
I designed and built it, including its agent runtime and terminal interface in Zig.
Case study
Handwork
What I focused on
For Handwork, I brought product thinking, interface design, and frontend engineering together around one clear problem.
Handwork connects a language model to a project through a native agent loop. The model can read files, search code, make targeted edits, run commands and tests, and report what changed.
The agent runtime and terminal interface are written in Zig. Handwork works with cloud model providers or a local Ollama server, saves and resumes conversations, delegates bounded tasks to subagents, and applies permission rules before tool actions.
A desktop app adds projects, attachments, and offline voice dictation. A JavaScript SDK lets other applications embed the agent with their own interface, tools, or authentication.
The site publishes coding task results against OpenCode, Codex, and Claude Code, including failed attempts, and states that those four tasks do not establish an overall ranking.
What I can help with
- Websites and web apps
Custom builds shaped around the product, audience, and way the team needs to work.
- Frontend systems
Responsive layouts, reusable components, semantic HTML, and architecture that is easy to extend.
- Interaction and motion
Purposeful animation, 3D experiences, and polished details that make an interface feel intuitive.
- Product interfaces
Dashboards, learning tools, AI products, workflows, and design systems built for real use.
- Performance and accessibility
Fast, crawlable pages with clear content, structured data, and accessible interactions.
