I recently wrote an answer related to this, particularly around how to take skills in AI and translate them into an enterprise environment.
Anand Sampat's answer to Most AI practitioners have deep technical expertise and are comfortable in that environment, whereas adopting AI in the enterprise requires a significant level of business understanding. How can this gap be bridged?
The main point is that AI / deep learning skills is not necessarily enough for implementing algorithms in production unless you have a team that includes all of the necessarily components. In addition to AI / deep learning skills, you will need domain expertise, dev ops knowledge, an understanding of the business goals and how they translate to the metrics you care about.
There are many places you can go to learn about machine learning and AI online, but understanding the specific domain is more variable, but learning that is the difference between an algorithm that falls flat, and one that changes the game for the business.
Hope this helps :)