What we have built.
The public side of it, across implementation, orchestration and sovereign deployment: what the ask was, what turned out to be the real problem underneath it, and what the system does now.
An inbox that sorts by sender, not by arrival time.
“People lose an hour a day in here. Can it sort itself out?”
Grouping by sender across the mailbox, rules that raise and file unasked, and a model that learns what each person opens.
Productivity software · GenAI implementationAn assistant that prepares the meeting and files the record.
“Reps spend Monday reading about people and Friday typing up what they said.”
A brief assembled when a meeting is booked, an assistant that follows the call, and a record written from what was said.
B2B sales software · Agents and orchestrationProduct video from one photo and ten seconds of audio.
“A minute of finished product video costs us days, and none of that is the idea.”
An avatar built from one photo and ten seconds of audio, with B-roll and stills drawn from whichever model does that job best.
Creator and brand marketing · Agents and orchestrationOne item that is a document, a database and a graph.
“Every tool makes us pick a shape before we know what we are writing.”
One item holding blocks, types, properties and named links at once, shared in real time, with any view over it.
Knowledge softwareCompetitive tracking that stays current on its own.
“We priced this in a spreadsheet eighteen months ago and nobody has opened it since.”
Continuous reading of competitor pricing, changelogs and reviews, pulled into one comparable shape, with pricing models over the live market.
SaaS strategy tooling · Agents and orchestrationScreening that reads a career instead of matching a keyword.
“We are rejecting people we should be hiring, and we cannot tell you which.”
Screening that reads what somebody has done, evaluation scored against a written definition of the role, and the scheduling taken off the coordinator.
Recruitment · Agents and orchestrationFailure prediction and training, in one record that survives an audit.
“It has to predict the failure, and it has to survive the audit.”
Failure prediction on machine and operational logs, training that adapts and certifies, and one traceable record covering both.
Maintenance management and ed-tech · Sovereign AI