Own your data. Use AI on it.

What we build

We do the full build for you, in accounts that you control. We build three parts. Each part is useful alone. The details for small businesses

1. A data warehouse that you own We put your business data in one place, in your own cloud account. It uses an open-source database, such as Postgres.
2. Connectors for your SaaS tools Connectors pull data from the tools that you use now. They find the same customer, product, or job in each tool and match the records.
3. AI agents for your industry Agents draft, check, and flag work from that data. They use only the data and tools that you allow. They record each action, and a person approves the result.

Pick the task that you want to stop

We start with one task. Each page shows what we build for that task.

1. Check invoices An agent matches each invoice to the vendor and purchase order. It flags problems before you approve.
2. Send quotes the same day An agent writes draft quotes from your job notes and your past prices.
3. Reply to each inquiry An agent sorts each new inquiry, writes a draft reply, and records it in your CRM. A person sends it.
4. Get a morning brief One page shows what changed in your sales, jobs, and cash, and what needs you today.
5. Answer staff questions An agent answers from your procedures and past work. It shows the source of each answer.
6. Intake, drafts, and billing For professional firms: client intake, first drafts from your templates, and time checks before invoices go out.
7. Check documents against your rules An agent flags issues for a person to review. It records each check and each approval.
8. Not sure yet? Tell us about your week. We tell you which task is worth an agent, even if the answer is no.

Why we build it this way

A model on your own server can still lock you in. The lock can come from APIs, identity, proprietary data, or prompts that nobody documented. It can also come from agent runtimes, workflow engines, cloud infrastructure, or one provider that you cannot replace. An open-weight model on your server does not make you sovereign. A self-hosted application does not make you sovereign. Sovereignty is a property of the full system.

The architecture

Sovereign Source AI has seven layers that support each other.

1. 01 Intent What a person or institution wants to do. It is a typed object that links to requirements, constraints, authority, and outcomes that you can measure.
2. 02 Governance Authority, policy, data classification, jurisdiction, model and tool permissions, network access, and approvals. Authorization is outside the model. A model can propose an action. It cannot give itself permission.
3. 03 Decision Deterministic rules, bounded models, statistical inference, LLMs that reason, and human judgment.
4. 04 Capability You declare each operation and its permissions. The system makes them available as Sovereign Skills and MCP-style interfaces.
5. 05 Execution Isolated agents, workers, workflows, services, and infrastructure.
6. 06 Verification Tests, evaluations, policy checks, security analysis, simulation, approvals, and measurement of outcomes. These checks change claims into evidence.
7. 07 Evidence Durable provenance records that link intent to decisions, actions, artifacts, releases, and outcomes.

Topics

Ten short pages explain the architecture, its reference builds, and parts of the research.

1. 01 Architecture Overview Sovereign Source AI treats sovereignty as a property of the whole system that does your work.
2. 02 Intent and SSIL Intent records what an institution wants to do and why.
3. 03 Governance and Policy Governance sets who can act, what they can do, and which conditions must be true.
4. 04 Decision Plane The decision plane uses typed context from the institution to propose a choice, an assessment, or a plan.
5. 05 Capabilities and Sovereign Skills The capability layer lists the operations that an actor can do and the conditions for each operation.
6. 06 Execution and Runtime Execution turns an authorized work order into actions, artifacts, and recorded results.
7. 07 Verification and Evaluation Verification turns claims about a system into evidence that the institution can use for decisions.
8. 08 Evidence, Provenance, and Supply Chain Evidence lets an institution explain why work started, who approved it, what made each artifact, and what happened after deployment.
9. 09 Data, Ontology, and Events The sovereign ontology gives your institution one structure for its knowledge, and applications, models, and workflows all use it.
10. 10 Reference Stack and Exit A reference stack shows how the parts of Sovereign Source AI fit together.

What it is not

Not a model

It works with any LLM, any cloud, and any vendor, and you can replace each vendor. Palantir, Payload, PostHog, MCP servers, open models, proprietary frontier models, and other products can be parts of the system. None of them owns the core model of your institution.

Not a compliance checkbox

It is an operating layer for your institution. It turns your knowledge, policies, requirements, authority, capabilities, markets, software, evidence, and outcomes into typed objects. You can run and audit these objects.