// the find
campusx-official/ML-Roadmap-for-2022
A curated list of Machine learning videos, links, projects and datasets to help you conquer the ML landscape in 6 months
A structured ML learning roadmap from CampusX, an Indian EdTech YouTube channel. It's a sequenced playlist aggregator — Python basics through model deployment — aimed at complete beginners who need someone to tell them what to watch and in what order. Not a course, not code, just a curated path through YouTube.
The five-level progression is sensible and the time estimates are honest enough to be useful. Coverage of practical preprocessing topics (missing values, encoding, outliers, imbalanced data) is more thorough than most roadmaps that skip straight to algorithms. The Kaggle notebook links for EDA practice are genuinely useful for building intuition on real datasets.
Everything links to one YouTube channel (CampusX's own), which means you're getting one pedagogical voice for every topic — if their explanation of, say, gradient boosting doesn't click for you, you're on your own. The roadmap stops at 2022 and is explicitly frozen: NLP section references pre-transformer techniques, Computer Vision and Time Series are marked 'Coming Soon' with no content. Heroku deployment links are dead — Heroku killed free tiers in 2022. No LLMs, no modern MLOps tooling (MLflow, DVC, etc.), nothing about vector databases or embeddings, so anyone following this in 2025 will finish it and still be years behind industry practice.