LoomWerk · a Southmead product
Your AI Virtual Office for a Team of Agents
LoomWerk is the AI Virtual Office: run an AI company inside your company. Describe an operation in natural language and it writes the playbook, staffs AI roles, and runs the work on your server. Management sets the rules; AI staff work within them.
HOW IT WORKS
From a description to running work
Four steps, in natural language from start to finish.
-
Describe
You describe the operation in natural language.
-
Playbook
LoomWerk writes the playbook: readable, editable steps your team owns.
-
Staff
It staffs AI roles, each with a job description and rules, and runs the work on your server.
-
Manage
Management sets the rules; AI staff work within them.
THE ENGINE
The engine checks itself before it acts
Its name is Quantum Trust Thinking (QTT™): a proprietary algorithm leveraging MoE with confidence scoring to verify truth.
Scrambled. Two phases to go.
Before LoomWerk's engine acts, three internal roles, strategy, structure, logic, have to agree. Anything below the certainty bar pauses and asks instead of guessing.
Outputs report their confidence, and execution is blocked when confidence is too low or information is missing.
Unsure means stop, not bluff.
TRUST
Checks before anything runs
People stay in charge. Unsure output stops; hard cases reach a person with the full history.
Built-in checks
Every planned output risk-scored 0–1; low passes, medium passes with warnings, high is corrected and retried.
Where do people fit?
At most 3 correction rounds, then human escalation with the full history attached.
What did it actually do?
Escalation alerts by email or webhook the moment they happen. Every escalation carries the full history, and the platform keeps an activity log.
AI staff
AI roles with a job description and rules. They work through your tools, never directly with each other.
WHERE IT RUNS
Your server, your choice of AI
Runs on the client's own/public server; data stays home.
Your server
a small server, no special hardware. Per-staff API keys in an encrypted vault, masked on screen.
Cloud computing mode
Cloud computing mode: only the AI calls you configure leave your server.
On-premises mode
On-premises mode: nothing leaves your server. LoomWerk does not train on client data.
We can walk your IT lead through the setup any time.
LICENCE
Your decision
Then a licence scoped to what runs.
Pricing is scoped per deployment; subscription and one-time models are planned. Write to us at hello@southmead.asia and we'll come back to you with current pilot terms.
FAQ
Questions people ask us
How does the AI decide?
Behind it is Quantum Trust Thinking (QTT™), our proprietary algorithm leveraging MoE with confidence scoring to verify truth. When it isn't sure, it stops and asks a person instead of guessing: and the hard calls reach a person with the full history attached.
What if it makes mistakes?
We don't promise zero mistakes: nobody honestly can. We promise every output is checked before it runs, unusual cases stop and escalate instead of running blind, and the hard calls reach a person with the full history attached.
When do you launch?
Launch timing is Southmead's to announce, and pilot clients hear it first. The way in today is the pilot: one workflow on your own hardware, full logs, then your decision.
Does it work for our industry?
Built for the operation, not the industry. The same playbooks, gates and escalation paths run back-office work wherever a team can describe it. The pilot is the offer, and pilot clients get a direct say in the roadmap. One workflow, full logs, then your decision.
THE PILOT
Start with a pilot. Decide on evidence.
The pilot is the offer, and pilot clients get a direct say in the roadmap. One workflow, full logs, then your decision.