AI & Automation

My AI work focuses on connecting models to useful tools and workflows, not training foundation models. Through personal projects and experiments, I explore coding agents, persistent memory, integrations, and automation.

My contribution is the integration: connecting tools, designing workflows, managing context, and validating results. The underlying models and frameworks are existing tools, not my own inventions.

Example: a connected personal assistant

SELECT * FROM personal_projects WHERE topic = 'assistant'

One personal project has explored how an assistant can combine conversation, long-term context, external tools, and recurring tasks. Examples include preparing email drafts, producing source-linked digests, recalling project decisions, and answering voice queries.

It was built from existing open-source parts such as Hermes Agent for the assistant and Hindsight for memory, with Home Assistant for the voice side.

An earlier experiment, mai (said like “my”, as in “my AI”), was a proof of concept of an “AI operating system”: a full-screen desktop app where a fast model answers and hands real work to background Claude agents with shell, file and browser access. Risky steps such as deleting files or pushing code wait on an approval card, and every job ends with a short report. Built with Tauri, TypeScript and the Claude Agent SDK; no longer actively developed.

The general design, simplified. The last step is a principle I design for, not a single technical gate.

Areas of experience

SELECT area, summary FROM ai_experience

Agent-assisted software development

Splitting work into implementation, tests and review, with review done by an agent that didn’t write the code. Handing context between agents through written briefs and shared rules, and letting a stronger model plan while simpler tasks go to cheaper ones. For the human side I built verifair, where agents publish change reports with their evidence and people sign, comment on and approve them instead of reading every diff.

Memory & personalization

Extracting facts, recalling context and consolidating memory, while keeping where each fact came from and telling confirmed information apart from guesses and outdated details.

Tool integration

Connecting models to APIs, MCP servers, the web and real browsers, and to productivity and home systems.

Research & productivity

Source-linked digests, product comparisons, keeping track of information, draft replies and follow-up tasks.

Voice & home automation

Experiments with voice interfaces, keeping simple local commands deterministic and sending only harder questions to an AI model.

Evaluation & reliability

Comparing models and agents on concrete tasks by quality, speed, cost and reliability. Experiments have included agents that summarize alerts with a likely cause, for a person to decide on, plus checking results, monitoring and making sure things can be restored.

Principles & lessons

SELECT lesson FROM experience

I prefer limited permissions, human review for sensitive actions, and verification against real systems. For simple or critical operations, deterministic automation is often the better choice.