I'm a third-year AI/ML student who has learned mostly through self-study, while also managing a demanding college schedule. So far, I've covered Python, DBMS and basic SQL, Flask, HTML/CSS/Bootstrap, and several DSA topics including arrays, hash maps and hash sets, linked lists, trees, graphs, heaps, and Dijkstra's algorithm. I'm currently learning tries and recently built a task manager API with FastAPI. I'm considering building another project using FastAPI, WebSockets, and PostgreSQL to learn more about real-time backend development.
My main challenge is deciding how to use my limited time. I don't want to restart DSA from the beginning, but I also want to make sure gaps in my fundamentals don't hurt me during interviews. I'm looking for advice from students who have balanced college with self-directed learning:
1. Is it better to continue learning new DSA topics while regularly revising older ones, or pause and review everything first?
2. How can I schedule DSA practice around a heavy college workload?
3. How can I use AI as a learning aid without becoming dependent on generated solutions?
4. If I were preparing for internships over the next few months, how should I divide my time between DSA, computer science fundamentals, backend development, and ML/DL?
My ML experience is still at the beginner level, so I'm also working on Python, mathematics, and general foundations while building practical projects.
2 Answers
You probably shouldn't restart DSA from the beginning. Keep moving forward with new material, but schedule short revision sessions during the week. For example, spend 30–45 minutes a few times weekly re-solving older problems from scratch, especially the ones where you needed hints or made mistakes. A mistake log is much more useful than repeatedly reviewing topics you already know.
With a busy schedule, protect a fixed DSA slot of roughly 45 minutes three or four times per week. Consistency is more valuable than occasional marathon sessions. For internship preparation, give DSA the largest share of your time, backend development the next-largest share, and use your coursework to support the ML/DL portion. A FastAPI, WebSockets, and PostgreSQL project is a worthwhile choice because it can teach connection management, state, and concurrency while giving you something concrete to discuss in interviews.
When using AI, attempt each problem yourself first. If you get stuck, ask for a hint, an explanation of the relevant concept, or a review of your approach instead of requesting the complete solution. After seeing an explanation, close it and implement the idea again from memory later. If you can't reproduce it, you likely need more practice.
Since your ML experience is still at the introductory stage, it would be useful to strengthen the mathematical foundations behind ML, such as linear algebra, probability, statistics, and basic calculus. The right balance depends on your internship target: ML-focused roles will require more math and model-building, while general software roles will usually place more emphasis on DSA and CS fundamentals.
You don't necessarily need to abandon backend work, though. A practical project can help you learn engineering skills and demonstrate that you can build something complete. Just avoid spreading yourself so thin that none of your areas gets enough focused practice.
I'm currently still learning the basics of ML, including the main learning types and why ML is used, so I haven't gone deeply into algorithms or the mathematics yet. For now, I'm focusing on Python, DSA, math, and practical projects before moving into more advanced ML topics.
That makes sense. I was overthinking the idea of restarting DSA, so I'll keep progressing while revising consistently and use AI mainly for hints and explanations.