The engine. Donna is model-agnostic, selecting the right model for each task rather than forcing every job through one. As the models improve, Donna improves with them, without the firm having to change how it works.

The control system. An LLM can produce an answer. It cannot run legal work. The harness plans the job, calls the right tools, follows the steps and asks for approval. It is what turns general-purpose intelligence into dependable legal agents.

The matter. Documents, emails, messages and practice-management data flow into each Space as working material. Native integrations put Donna inside the firm’s existing systems, so it works from the whole matter, not from one document in isolation.

The memory. Donna retrieves from the matter, cites the exact source, keeps the Space current and draws on the firm’s precedents, playbooks and clause libraries. Every action carries the facts of the matter and the firm’s accumulated judgment.

The work a firm recognises. Drafting from your own precedents. Reviewing and comparing documents. Building timelines. Running workflow agents. Handling multi-step requests through Concierge. Donna does not stop at an answer. It carries the matter forward.

The place. One Space per matter, where the firm, the client and Donna work together. Everything the platform puts in front of you lives here, so every action stays visible, editable and connected to the matter.

The safeguards. Security is not bolted on at the end. Identity, permissions, cross-matter isolation, private networking and audit trails govern every action beneath the model. Donna never trains on your matters.