This is really cool. I don't think you're aware of how valuable your Bayesian combination is. The reason I could claim that is that your first point was this:

This is really cool. I don't think you're aware of how valuable your Bayesian combination is. The reason I could claim that is that your first point was this: >1. What should I focus on learning to get a backend developer job that pays 180k+ in a major market You could make more as a developer; you could make more as a lawyer, but you're a develawyer (or a lawyeroper) and a non-linearity should apply. I have some ideas for your profile. First, the obvious ones: https://www.avodocs.com https://www.avodocs.com (YC) does legal documents for startups. You could work one the programing and the legal aspects for the crowd you seem predisposed to serve given your presence in this forum. https://rebase.co https://rebase.co. Description : >Rebase is a new, safe, and easy-to-use platform by Nomad List for becoming a legal and fiscal resident in countries that want to attract remote workers It sits between users and lawyers to manage the full process. It started this year and I think you could link up with https://twitter.com/levelsio https://twitter.com/levelsio and see what you could do to scale it. Again, whether you wear your developer hat and lawyer glasses, or your lawyer hat and developer glasses; the results will be interesting. Now for the less "obvious" things. I work at a boutique consultancy that builds end to end machine learning products for large organizations. Machine learning means we work with data and working with data means a whole lot of questions that have to be answered. Compliance with a law that was just passed or a legislation project that may become law in the future. Sometimes, you work with an organization that serves clients, and these clients have data of their clients. The project means you may have to use data of your client A, which is data of their clients B, C, D, which itself is data of the nested "clients" E, F, G. The conversation becomes: whose data is this ? What can you do ? What can't you do ? What should the client transmit ? Everything ? Out of the question. Is the client very sensitive (health, defense, finance, communication)? Where will the models be trained and what's the data flow (these are the meetings where people talk about territory, sovereignty, and borders). Where will the software be deployed. Which cloud provider, in which country. On-premises ? How to maintain and who will access from where ? There are many people involved. Many conversations. Many consultations. Mutual availability. Meetings scheduled weeks or months in advance. How about you productize this for these problems that pretty much every organization that wants to do anything with data has meetings about. No need to re-invent the wheel. You make a cookie-cutter. One way to do this is to join a large company that does this kind of projects for organizations. Learn what the friction points are, then develop a product (software + legal) to do that.