The problem it solves
A new hire usually gets a stack of documents: processes, product information, internal rules. Reading them is not the same as knowing them, so an experienced colleague ends up explaining the same things again, for every new person. Turning the documents into a proper course would help, but building one takes time a small company rarely has.
What it does
The company uploads its own documents: processes, product information, internal rules. The AI turns them into voice-based training sessions for the new hire. There is no manual course creation. The material comes from the document itself, and the pace adapts to what the person has already mastered.
For the company this means the knowledge it has already written down becomes something a new person can practise, in conversation, instead of something they are told to read.
How it is built
It runs on the same engine as the interview preparation case: a voice-first conversational AI that holds realistic practice sessions. Under the hood, every line the AI says is checked against a live model of what the person already knows. That keeps the conversation at the right level and introduces one new thing at a time, which matters in onboarding more than anywhere, because the first week is exactly when people are overloaded. Session history drives what comes next.
Privacy by design
Internal documents are company knowledge, and a new hire's learning progress is personal data. I design this kind of system the way I design every other: the AI receives only the parts of the documents it needs for the current session, and raw input is separated from the result. Uploaded documents stay with the company.
What it shows about how I build
The same engine serves two scenarios, a candidate preparing for an interview and a company training its new people. That is how I think about AI in operations: the valuable part is not the model, it is the mechanism around it, here a live picture of what a person knows, which decides what the AI says next. The company does not write a course; the system works from material that already exists. And the conversation is spoken because that is how people learn a job from a colleague.