If you say "is hiring me" I assume you've not started yet. Given that launching a startup is no joke, if I were you I'd start working on the tool for them, paid

If you say "is hiring me" I assume you've not started yet. Given that launching a startup is no joke, if I were you I'd start working on the tool for them, paid in full, keeping your idea to turn it into a startup in background. After a bit, if you're still committed to that, you feel you have cofounder(s) and money, then you can start sharing the idea with the client. I personally very much doubt switching from internal tool to SaaS will be no brain, and as long as you don't reuse their code you should be free to rebuild a similar SaaS. Test their interest in becoming your first customer and/or an investor in your startup (perhaps in exchange of the code you already wrote for them). I'd be very cautious because if there are alternatives that are quite expensive... there might be a reason for that. And I wouldn't be too optimistic about your personal cost (time & money) to switch an internal project into a startup. This said, best of luck, and post it on HN when you launch!