AI/ML roadmap: from Python to production models
For CSE (AI/ML) students: maths foundations, classical ML, deep learning, MLOps and a portfolio that proves it.
4 stages · 18 steps. Open any step to see why it matters, what comes before it and where to learn it.
Years are a guide, not a deadline. Students move at different speeds and in a different order; skip what you already know and spend longer where you need to. Steps marked Optional are worth doing if they fit your goals.
First year
Python, maths and curiosity.
Second year
Classical ML and honest evaluation.
Third year
Deep learning and a specialisation.
Fourth year
MLOps, interviews and the story of your work.
This is guidance written by the UniCircle team, not a rulebook, and it promises no outcome. Pair it with the live openings, the courses and interview practice, and ask seniors in your college community what actually worked for them.
Reference: we checked the skill order against roadmap.sh/ai-data-scientist. The steps and wording here are our own; linked resources belong to their publishers.