I've been learning C++, data structures and algorithms, and ML for about eight months. I've tried implementing algorithms, built some projects, and explored deep learning and AI agents, but now I'm overwhelmed by all the advice about what to learn next.
I enjoy ML more than DSA, though I know I still have gaps. Should I focus on MLOps, go deeper into ML, or try something else?
Pick one direction for a while rather than trying to settle your whole career right now. You could take an ML project you've already built and work on understanding it more deeply: explain the data, the model, how you evaluated it, and where it fails.
Revisit the concepts you can't explain or change without following a video. Eight months is enough to explore a lot, but there's no need to expect yourself to have mastered every topic you've touched.
MLOps is worth exploring if you like the engineering around models, not just training them. Try packaging one existing model so it can accept new input, return a prediction, and run consistently in a clean environment.
That gives you a small taste of deployment and reproducibility without committing to a whole new stack. If you enjoy that more than experimenting with models, follow it further. If not, it's fine to stay closer to ML.
Cut back on direction videos for a bit. Pick a project question that interests you and build a small version without following a complete tutorial. Use documentation or targeted explanations when you get stuck, then go back to the project.
It doesn't have to be an impressive new agent system. Improving a previous project is a perfectly good way to find out what you want to learn next.
Keep DSA on a smaller, steady schedule instead of making it your entire focus. Review the fundamentals you haven't finished and practice applying them, while spending most of your project time on ML if that's what holds your interest.
How much interview-style practice you need will depend on the roles you eventually apply for. You don't have to decide that from a video about someone else's career.
Answers are reviewed before publication.
hAlF+MARGIN:
That also makes it easier to see whether you enjoy the model work itself or mostly enjoyed trying out new tools.