Answers
How do you onboard a new hire into a team that already works with AI agents?
Give the new person, on day one, the four things they would otherwise learn by interrupting people: which agents the team uses and who answers for each, what has been decided and is in force today, what is half-done and with whom, and how to ask for a correction when an agent gets something wrong. Harvard Business Review's March 2026 advice treats agents as colleagues who need an onboarding plan of their own; the other side of that is that the human arriving beside them needs one too, and it should not be the full history of the project. The shortest version is a read, not a document: the new person's own assistant opens by reading what is in force on the project and tells them. Arroway is where that lives: decisions with who made them and when they end, handoffs with an owner, and the log of what each agent did, read by every connected tool before it acts. The install page is the first step.
Last updated October 6, 2026
What the new person needs on day one
Four things, in this order. Which agents the team works with, and the person who answers for each one. What is decided and still valid: the pricing rule, the client's constraint, the thing the team tried and dropped. What is in flight: work an agent or a colleague left half-done, and with whom it sits. And how to correct an agent, meaning whom to tell and where the correction goes so the next session does not repeat the mistake. The full history is not on the list. A new hire who gets six months of chat and files has the same problem as an agent given the whole context window: the current rule is in there somewhere.
The first week without interrupting anyone
When this lives in people's heads, the new person learns it by asking, and the answer depends on whom they ask. When it lives in a shared document, it is as old as its last edit. When it lives where the team's assistants read it before they act, the new person's own assistant can answer the same questions from the same source: what is in force, who decided it, what replaced what. Their corrections go back as proposals, and a person approves what becomes a rule. This is also what keeps the agents consistent with each other while someone new is learning the ropes.
When to use Arroway
Use Arroway when the team has more than one person and more than one AI, and the answer to what is decided should not depend on who is asked. Each decision carries who made it and the condition that ends it; each handoff names who it is for; every connected assistant reads the current version at the start of a session. It does not replace a manager, a buddy or a first-day conversation. What it removes is the part of onboarding that is only transferring what the team already knows. The first step is the install page.
What it looks like in practice
A twelve-person agency hires an account coordinator in November. The team runs a research agent, a drafting agent and a reporting agent, used by different people in Claude, ChatGPT and Cursor. Before she starts, her future manager has nothing to write: the project already holds what is in force. Never send a client report without the account lead's approval. The drafting agent writes in the client's language but not in their legal register. The reporting agent was moved to weekly in October, replacing the monthly rule. On her first morning she opens her own assistant and asks what she needs to know about the Hartley account. It reads the project and tells her the three rules, the handoff left half-done on Friday and the name of the person who answers for each agent. When she spots that the drafting agent still uses an old tagline, she records the correction as a proposal; the account lead approves it that afternoon, and every assistant on the team reads the new line the next day.
Questions people ask about this
- Does the new hire need to read the whole project history?
- No. They need what is in force today, what is in flight and who answers for what. The history is there for the day they want to know why something changed, and it stays out of the first read.
- Does Arroway replace the onboarding conversation?
- No. A person still introduces the team and the way it works. Arroway covers the part that is only passing on decisions the team already made, so the conversation can go to judgment and not to repeating the rules.
- How do I try Arroway?
- Install it in the tool you already use — the install page has one route for Claude, ChatGPT, Codex, Cursor or any MCP tool — and sign in when the tool asks. From then on, sessions open by reading the project and close by recording what was decided.
Where this is verifiable
Harvard Business Review, 'Create an Onboarding Plan for AI Agents' (Joseph B. Fuller, 25 March 2026), for the idea that agents joining a team need an onboarding plan; the four-item list for the new person is this page's own working template, not the article's. Product documentation on this site (How it works, Install), the answers on who is accountable for an AI agent on a team and on handoffs between partners on the same account, and the sanction, supersession and expiry-condition rules in the sanctioned product spec. Everything described here about Arroway is behaviour the tools apply today, not roadmap. The agency in the example is illustrative.
https://www.arroway.app/en/answers/onboarding-a-new-hire-into-a-team-that-works-with-ai-agents