The AI team
Six roles, how to hire them, the tools they have, the method they follow, how work is tracked and checked, and what the team remembers.
- The six AI employee roles: Researcher, Outreach (SDR), Writer, Engineer, Analyst and Operator. Each starts from a role template with a working style, a tool set and the connectors it needs.
- Hiring and managing employees: Hire from a role, rename them, edit how they work and what they can use, pause them, set autopilot per app, or let them go. Limits come from your plan.
- Tools reference: Every tool an employee can be given, which run on its computer, which need an app, and which wait for your approval. Plus custom tools the team can build.
- Custom tools: Build a tool once and every agent can call it: a script, a public web API, or a prompt template. Agents can build them too. You can switch one off, read its code or delete it.
- How employees do a job: Every employee follows the same method: understand, plan, do, check, report. Each role has a playbook and each kind of job has its own report shape and step budget.
- Reports and delivery: Each kind of job has its own report shape. The server, not the model, decides whether the thing you asked for was delivered, and says so plainly.
- Quality controls: Failures return a recovery hint, a call that failed twice is not run a third time, substantial work is reviewed once by its author, and the checker counts what it can count.
- Tracked tasks: A tracked task has a brief, a definition of done, its own channel and a progress log. Its work is checked against the definition, sent back to improve, then handed to you.
- Runs: how work executes: Every piece of agent work is a durable run: one leased turn at a time, resumed if a worker dies, continued in parts, and recoverable without anyone having the app open.
- Group chats and hand-offs: Put several employees in one room, mention them by name, and let them hand work to each other with a typed brief. Limits keep it from spiralling.
- Shared memory: One memory shared by Pytheus and every employee: facts, preferences, decisions and findings. Agents read it before they start. You can add, pin and forget entries.