I would say that you should build the MVP, as a side project only. Take care that what you do won't use any company resource to build. Don't quit until you are

I would say that you should build the MVP, as a side project only. Take care that what you do won't use any company resource to build. Don't quit until you are sure this will work, or you are going to burn a very well paid bridge. Unless you are building a deep-tech product (blockchain, AI stuffs), try to put together your project using pre-made parts. I'd recommend taking a look at Low-code/No-code tool. When you have validated your product against the market, you can then quit and focus full time on the project. I have personally help a few startup firms build their MVP using those pre-made parts to validate their idea. Some of them works, but for those failed, they didn't waste too much time and money on it. Those ideas that work, they quickly turned to a full-fledged solution and ditched the MVP. About the legal aspect, I think that unless the contract really specifies the IP laws, I think it's gonna be alright. Maybe consult a Lawyer about it?