Learning Roadmap. Visual Guide

A visual representation of your ML learning journey from zero to hero.

Numbering: Stages below are the recommended learning order. Module numbers (00–25) are folder names on disk. Module 09 is not Stage 9. See START-HERE.md.

Job-market fix: For analyst and data scientist roles, treat Module 19 (SQL) as Stage 1.5, parallel with Stage 1–2, not after GenAI. Full exit gates: FOUNDATION_AND_JOB_READINESS.md.

Learning Path Overview


                    ROAD TO ML: ZERO TO HERO                  


STAGE 0: FOUNDATION (2-3 months full-time, 4-6 months part-time)
 Python Basics
 Mathematics Basics
 Environment Setup
    ↓
STAGE 1: DATA FUNDAMENTALS (2-3 months full-time, 4-6 months part-time)
 NumPy, Pandas & Polars
 Data Visualization
 Exploratory Data Analysis
    ↓
STAGE 1.5: SQL & DATABASES (parallel with Stage 1–2 for job tracks)
 Module 19, queries, joins, window functions
    ↓
STAGE 2: ML BASICS (2-3 months full-time, 4-6 months part-time)
 Introduction to ML
 Regression & Classification
 Bias Auditing & Fairness
 Model Evaluation
    ↓
STAGE 3: ADVANCED ML (1-2 months full-time, 2-4 months part-time)
 Ensemble Methods
 Feature Engineering
    ↓
STAGE 4: UNSUPERVISED LEARNING (1 month full-time, 2 months part-time)
 Clustering
 Dimensionality Reduction
    ↓
STAGE 5: DEEP LEARNING FUNDAMENTALS (2 months full-time, 4 months part-time)
 Neural Networks Basics
 Deep Learning Frameworks (PyTorch/TensorFlow)
    ↓
STAGE 6: SPECIALIZED DEEP LEARNING (3-4 months full-time, 6-8 months part-time)
 Computer Vision
 Natural Language Processing
 Time Series (Module 15 OR project, see TIME_SERIES_LEARNING_PATH.md)
    ↓
STAGE 7: GENERATIVE AI & MODERN LLMs (1-2 months full-time, 2-4 months part-time)
 Prompt Engineering
 RAG & Vector Databases
 LLM Agents & Multi-Agent Systems
    ↓
STAGE 7.5: ESSENTIAL SKILLS (1-2 months full-time, 2-4 months part-time)
 Imbalanced Data (Module 20)
 Model Explainability (Module 21)
 DSA in Python (resources/dsa_course_python.md, interview prep)
 Note: SQL is Stage 1.5 (Module 19), not here
    ↓
STAGE 8: PRODUCTION & MLOPS (2-3 months full-time, 4-6 months part-time)
 Model Deployment (API, Docker, Cloud)
 Model Optimization (Quantization, ONNX)
 MLOps Basics (MLflow, W&B)
    ↓
STAGE 9: PROJECTS (Ongoing)
 Beginner, Intermediate & Advanced Projects
    ↓
STAGE 10: ADVANCED SPECIALIZATION (2-3 months full-time)
 Reinforcement Learning
 Graph Neural Networks
 Audio Processing

Job market alignment

Default stage order is comprehensive but not optimal for every target role. Use FOUNDATION_AND_JOB_READINESS.md for:

Exit gates after Modules 00, 01, and 02–05
SQL early (Module 19 after 01) for Data Analyst and Data Scientist paths
Parallel projects starting after Stage 2
Portfolio and interview checklists

Milestone Checklist

Foundation Milestone (2-3 months)

Python basics mastered
Math concepts understood (with computational examples)
Environment set up
First Jupyter notebook created

Data Science Milestone (2-3 months)

Can manipulate data with Pandas
Understand when to use Polars/Dask for large datasets
Can create visualizations
Completed first data analysis

ML Basics Milestone (2-3 months)

Built first regression model
Built first classification model
Performed bias auditing on models
Understand model evaluation
Completed first ML project

Advanced ML Milestone (1–2 months after ML basics)

Used ensemble methods
Performed feature engineering
Applied clustering
Completed intermediate project

Deep Learning Milestone (2 months)

Built neural network
Trained CNN for images
Built NLP model
Completed deep learning project

Generative AI Milestone (1-2 months)

Understand prompt engineering
Built RAG application
Worked with vector databases
Created LLM agent

Production Milestone (2-3 months)

Deployed a model (including Hugging Face Spaces)
Optimized model with quantization
Set up MLOps pipeline (MLflow, W&B)
Completed advanced project
Portfolio ready!

Skill Progression

Beginner Level (after Stage 2)

Skills:

Python programming
Data manipulation
Basic ML algorithms
Model evaluation

Projects:

House price prediction
Iris classification
Titanic survival

Intermediate Level (after Stages 3–4)

Skills:

Feature engineering
Ensemble methods
Model optimization
Unsupervised learning

Projects:

Customer churn
Recommendation system
Fraud detection

Advanced Level (after Stages 5–8)

Skills:

Deep learning
Computer vision
NLP
Model deployment
MLOps

Projects:

Image classification
Sentiment analysis
Chatbot
End-to-end pipeline

Learning Strategies

For Complete Beginners

Don't Skip Basics: Foundation is crucial
Practice Daily: Consistency beats intensity
Build Projects: Apply what you learn
Ask Questions: Join communities
Be Patient: ML takes time to master

For Programmers New to ML

Review Data Science: NumPy, Pandas are essential
Focus on ML Concepts: Algorithms, evaluation
Build Projects Early: Learn by doing
Study Math: Understanding helps a lot

For Quick Learners

Skip What You Know: Focus on gaps
Challenge Yourself: Try advanced projects
Contribute: Help others learn
Specialize: Pick CV, NLP, or MLOps

Time Estimates

Note: These are realistic estimates for comprehensive learning. The "Accelerated Path" below is for experienced programmers only.

STAGE	Full-Time (30-40 hrs/week)	Part-Time (10-15 hrs/week)
Foundation	2-3 months	4-6 months
Data Science	2-3 months	4-6 months
ML Basics	2-3 months	4-6 months
Advanced ML	1-2 months	2-4 months
Deep Learning	2 months	4 months
Generative AI	1-2 months	2-4 months
Production	2-3 months	4-6 months
Projects	4-6 months	8-12 months

Total Time to Hero: 14-20 months (full-time) or 28-35 months (part-time) for comprehensive coverage

Quick Wins

Week 1

Set up environment
Write first Python program
Create first visualization

Week 2

Load and explore a dataset
Clean data
Create a simple model

Week 4

Complete first ML project
Deploy model locally
Share on GitHub

Accelerated Path (For Experienced Programmers)

If you already know Python and basic math:

Week 1: Data science basics (NumPy, Pandas)
Week 2: ML basics (Regression, Classification)
Week 3: Advanced ML (Ensembles, Feature Engineering)
Week 4: Deep Learning basics
Week 5-6: Specialize (CV or NLP)
Week 7-8: Build portfolio projects

Total: 2 months to advanced level

Resource Allocation

70% Practice

Coding exercises
Building projects
Experimenting

20% Learning

Reading documentation
Watching tutorials
Reading papers

10% Review

Revisiting concepts
Explaining to others
Teaching

Certification Path

While this repository doesn't offer certificates, you can:

Build Portfolio: Complete projects
Share on GitHub: Show your work
Write Blog Posts: Explain what you learned
Report issues: Broken links or errors via GitHub Issues on this repo
Get Certified: Consider Coursera, Udacity certificates

Staying Motivated

Set Goals: Weekly and monthly targets
Track Progress: Check off completed modules
Join Community: Find study partners
Celebrate Wins: Acknowledge progress
Take Breaks: Avoid burnout

Iterative Learning

Learn → Practice → Build → Review → Improve
  ↑                                    ↓

Learn: Study concepts
Practice: Do exercises
Build: Create projects
Review: Reflect on what you learned
Improve: Fill gaps, go deeper

Current Position Tracker

Use this to track where you are:

STAGE 0: Foundation
STAGE 1: Data Fundamentals
STAGE 1.5: SQL & Databases (Module 19, parallel if on analyst/DS track)
STAGE 2: ML Basics
STAGE 3: Advanced ML
STAGE 4: Unsupervised Learning
STAGE 5: Deep Learning Fundamentals
STAGE 6: Specialized Deep Learning
STAGE 7: Generative AI & Modern LLMs
STAGE 7.5: Essential Skills (modules 20–21; SQL is Stage 1.5)
STAGE 8: Production & MLOps
STAGE 9: Projects
STAGE 10: Advanced Specialized Topics

Next Steps

Assess Your Level: Where are you now?
Set Your Goal: Where do you want to be?
Plan Your Path: Which modules do you need?
Start Learning: Begin with the first module
Track Progress: Update this roadmap

Try next: Open Module 00 and finish one exercise block before you close the laptop.