Most people picture AI as a chat box that answers questions and forgets you the moment you close the tab. I build something more ambitious: complete AI systems that carry memory across time, make their own decisions, and run privately on your own computer. This site explains what that means — no jargon required.
Think of an AI model (like ChatGPT) as an engine. On its own, an engine just sits there. My job is to build the entire car around it — the memory, the controls, the safety, the wiring — so it becomes something genuinely useful and dependable.
Ordinary chatbots forget everything between conversations. I design memory systems so an AI can remember who you are, what matters, and what it learned last week.
Instead of only answering questions, my systems can plan multi-step tasks, use tools, and carry work forward without being babysat — safely and within limits you set.
Many of my systems run entirely on your own machine — nothing sent to the cloud — and keep a clear record of every action, so you can always see what happened and why.
Different problems need different shapes. Here's the spread of what I design — each one explained in a single plain sentence.
An assistant that's truly yours — it remembers your projects, works across chat, Discord and the web, and never phones home unless you let it.
Systems that notice a gap in their own abilities, build a new tool to fill it, test it safely, and keep it if it works — getting more capable over time.
Virtual towns and 3D worlds populated by characters that have goals, moods and memories — useful for research, games, training, and storytelling.
A "panel" of AI specialists — a CEO, a lawyer, an engineer — that discuss your problem out loud and hand you a rounded answer instead of one flat reply.
Instead of a mysterious black box, an AI whose knowledge is stored as plain, readable files you can open, check, and correct.
Drag-and-drop builders and ready-made starter kits so ideas can become working AI software far faster.
Your data is yours. I favour AI that runs locally, so sensitive information never has to leave your device.
You should be able to see what the AI knows and why it did what it did. I build in evidence trails and audit logs.
Powerful AI needs guardrails. Nothing important happens without your approval and clear limits.
Whether you're a business exploring AI or a fellow builder, I'm happy to talk through what's possible — in plain language.