Four interactive demos that run in your browser on sample data. They are not client projects. Each one shows how we build a piece of the work and what a partner project would add.
Staff lose time hunting through handbooks and policies. The partner needs a retrieval layer that cites its sources and says so when it cannot answer.
For: Software houses whose client has a pile of documents nobody can search.
Open demoPlanners need a forecast they can question. The partner needs to know whether a model actually beats the method already in use.
For: Software houses building planning, retail or operations products.
Open demoA client wants AI in their support tool, but needs predictable output, a person in control of risky cases and a record of every decision.
For: Agencies and product studios adding an AI feature to an existing app.
Open demoA brief arrives that mentions AI. Before quoting, the partner needs to know which parts are RAG, LLM, ML or integration work, and what is still unknown.
For: Delivery leads and agency owners deciding whether to take on an AI brief.
Open demoThe retrieval, forecasting and scoring code in these demos is real and runs on your device. The data is a sample or synthetic, and the triage and scope tools use rules where a production build would use a language model. We say so on each page. They show how we work; they are not evidence of client results.
Other public code is on our GitHub organisation.
Share a short brief. We’ll tell you whether the work is a fit and how a collaboration could be scoped.