AI Roadmap Progress Tracker

10-month plan · 40 weeks · 5 portfolio projects

The complete AI Engineer roadmap

A free, opinionated 10-month learning path from foundations to a job-ready AI Engineer. Math, classical ML, deep learning, transformers from scratch, RAG, fine-tuning, agents, ML system design, and interview prep — broken into day-by-day tasks with hand-picked resources. Browse the whole plan below. Sign in to track your progress, streak, and analytics.

Browse the roadmap
  • · LLMs (40%)
  • · Backend / MLOps (27%)
  • · ML foundations (20%)
  • · Interview prep (14%)
  • · NLP (5%)
  • · Computer vision (4%)

Roadmap Topics

Week 1 — Linear Algebra Foundations
0 / 7·0%
101.1Day 1 — Vectors, dot product, vector addition
1.5hMUST-KNOW
101.2Day 2 — Linear combinations, span, basis vectors
1.5hMUST-KNOW
101.3Day 3 — Matrix transformations
1.5hMUST-KNOW
101.4Day 4 — Matrix multiplication, 3D transformations
1.5hMUST-KNOW
101.5Day 5 — Determinants, inverse matrices
1.5hMUST-KNOW
101.6Day 6 — Eigenvectors & eigenvalues, SVD intuition
3hMUST-KNOW
101.7Day 7 — Implement matmul, eigendecomposition by hand in NumPy
3hMUST-KNOW
Week 2 — Calculus + Probability for ML
0 / 7·0%
102.1Day 1 — Derivatives, geometric meaning
1.5hMUST-KNOW
102.2Day 2 — Chain rule (THIS IS BACKPROP)
1.5hMUST-KNOW
102.3Day 3 — Partial derivatives, gradients
1.5hMUST-KNOW
102.4Day 4 — Probability basics, random variables, expectation, variance
1.5hMUST-KNOW
102.5Day 5 — Distributions: Bernoulli, Gaussian, Categorical
1.5hMUST-KNOW
102.6Day 6 — Bayes Theorem + conditional probability
3hMUST-KNOW
102.7Day 7 — Cross-entropy, KL divergence
3hMUST-KNOW
Week 3 — Python ML Toolchain + Linux/Git Basics
0 / 7·0%
103.1Day 1 — Install uv (modern Python package manager) + Linux/Bash basics
1.5hMUST-KNOW
103.2Day 2 — NumPy deep dive: broadcasting, vectorization
1.5hMUST-KNOW
103.3Day 3 — Pandas: DataFrames, filtering, groupby, merge
1.5hMUST-KNOW
103.4Day 4 — Pandas continued: pivot, time series, missing data
1.5hMUST-KNOW
103.5Day 5 — Matplotlib + Seaborn
1.5hMUST-KNOW
103.6Day 6 — Git/GitHub: branches, commits, PRs, .gitignore
3hMUST-KNOW
103.7Day 7 — Set up: GitHub, W&B, HuggingFace, Kaggle accounts. Push first repo.
3hMUST-KNOW
Week 4 — Classical ML Part 1: Regression + Classification
0 / 7·0%
104.1Day 1 — What is ML? Supervised vs Unsupervised. Train/val/test splits
1.5hMUST-KNOW
104.2Day 2 — Linear Regression (single variable) — math + intuition
1.5hMUST-KNOW
104.3Day 3 — Multiple linear regression, gradient descent
1.5hMUST-KNOW
104.4Day 4 — Implement linear regression from scratch in NumPy
1.5hMUST-KNOW
104.5Day 5 — Logistic regression, sigmoid, binary classification
1.5hMUST-KNOW
104.6Day 6 — Logistic regression from scratch + cost derivation
3hMUST-KNOW
104.7Day 7 — Overfitting, regularization (L1/L2), bias-variance
3hMUST-KNOW
Week 5 — Classical ML Part 2: Trees + Ensembles
0 / 7·0%
105.1Day 1 — Decision Trees: entropy, Gini
1.5hMUST-KNOW
105.2Day 2 — Random Forests + bagging
1.5hMUST-KNOW
105.3Day 3 — Gradient Boosting intuition
1.5hMUST-KNOW
105.4Day 4 — XGBoost + LightGBM hands-on
1.5hMUST-KNOW
105.5Day 5 — KNN, Naive Bayes overview
1.5hMUST-KNOW
105.6Day 6 — K-Means clustering + PCA
3hMUST-KNOW
105.7Day 7 — Compare 5 algorithms on Titanic dataset, F1 scores
3hMUST-KNOW
Week 6 — Model Evaluation + Feature Engineering + MLflow
0 / 7·0%
106.1Day 1 — Why accuracy misleads. Precision, Recall, F1
1.5hMUST-KNOW
106.2Day 2 — ROC, AUC, PR curves
1.5hMUST-KNOW
106.3Day 3 — k-fold + stratified cross-validation
1.5hMUST-KNOW
106.4Day 4 — Data leakage: causes + prevention
1.5hMUST-KNOW
106.5Day 5 — Feature scaling, one-hot, target encoding
1.5hMUST-KNOW
106.6Day 6 — sklearn Pipelines + ColumnTransformer
3hMUST-KNOW
106.7Day 7 — MLflow: track experiments with params + metrics + artifacts
3hMUST-KNOW
Week 7 — PROJECT 1: End-to-End Tabular ML Pipeline
0 / 7·0%
107.1Day 1 — Pick dataset, EDA, document findings
1.5hMUST-KNOW
107.2Day 2 — Feature engineering: missing values, encoding, scaling
1.5hMUST-KNOW
107.3Day 3 — Build sklearn Pipeline with ColumnTransformer
1.5hMUST-KNOW
107.4Day 4 — Train 3 models: LogReg baseline, RF, XGBoost. Stratified k-fold CV
1.5hMUST-KNOW
107.5Day 5 — Hyperparameter tuning with GridSearchCV/RandomizedSearchCV
1.5hMUST-KNOW
107.6Day 6 — Log everything in MLflow. Write README
3hMUST-KNOW
107.7Day 7 — Clean GitHub structure: /data, /notebooks, /src, README.md, requirements.txt
3hMUST-KNOW
Week 8 — Backend Foundations: HTTP, FastAPI, Docker, Cloud Deployment
0 / 7·0%
108.1Day 1 — HTTP fundamentals: methods, status codes, headers, JSON
1.5hMUST-KNOW
108.2Day 2 — FastAPI Hello World, path/query params, async basics
1.5hMUST-KNOW
108.3Day 3 — FastAPI: Pydantic request/response models, validation
1.5hMUST-KNOW
108.4Day 4 — Serve Project 1 model via FastAPI: /predict, /health, async endpoint
1.5hMUST-KNOW
108.5Day 5 — Docker: images, containers, Dockerfile, multi-stage build
1.5hMUST-KNOW
108.6Day 6 — Dockerize FastAPI service + write docker-compose.yml + env vars (.env)
3hMUST-KNOW
108.7Day 7 — Deploy to Render or HuggingFace Spaces. Public URL. Add error handling middleware.
3hMUST-KNOW

Track your progress privately

Create a free account to check off daily tasks, build a streak, view analytics, and see weekly progress with a circular ring. Your progress is stored privately to your account.

Built by Aryav Chanduka · B.Tech CSE AI/ML, Manipal University Jaipur