About MetaWeaver
Knowledge you can actually trust with your work.
MetaWeaver was founded on a simple conviction: the knowledge that matters most to you is already sitting in your own documents — it just isn’t working for you yet. A chatbot trained on the internet has never read your files, and you can’t hand it the keys to your work. We’re building the tool we wanted for ourselves — one that turns your own documents into a private knowledge base you can question, search, and put to work, with you approving anything that matters.
What we stand for
These aren’t slogans — they are the decisions we’ve built into the product from the start.
Private by design
MetaWeaver works only from your own documents, and your files stay yours. We never train our models — or anyone else’s — on your data.
Bring your own key
You choose the AI provider and use your own key, so you pay only for storage and never get locked in.
You approve what matters
Anything that sends, changes, or deletes waits for your approval. Nothing happens behind your back, and every action is logged.
Built to be trusted
A clear record of what changed and when — and the option to self-host with a local model, so nothing ever leaves your machine.
Who’s building it
MetaWeaver is built by two founders who come at the same problem from opposite ends — decades of building large-scale, AI-driven systems in the real world, and world-leading research into the AI that makes them possible.
James Liu
Co-founder · CEO & Chief Architect
Over 30 years in the IT industry — most recently applying AI to large-scale data analysis in network security, with earlier work spanning healthcare, telecom, defence, and the internet. Holds a Bachelor of Engineering from Tsinghua University. MetaWeaver grew out of a career spent building systems people actually rely on, and the conviction that the tools you trust with your work look nothing like a generic chatbot.
F. Richard Yu
Co-founder · Chief Scientist
Professor at Carleton University and one of the world's most-cited researchers in AI, autonomous systems, and security — with over 800 papers, 60,000+ citations, and an h-index above 131. He is ranked #1 worldwide in deep reinforcement learning and mobile edge computing (ScholarGPS), a Clarivate Highly Cited Researcher, and a Stanford Top 2% Scientist. He is a Fellow of the Royal Society of Canada, the Canadian Academy of Engineering, the IEEE, and the IET, and a Member of Academia Europaea. His expertise in AI and secure systems anchors the foundations MetaWeaver is built on.
View his faculty profile →Get in touch
Have a question, or want to talk about an on-premise or dedicated enterprise deployment? We’d love to hear from you — reach us at contact@ajaxweaver.com or through the contact form.