Data Science Roadmap: A 12-Week Plan From Zero to Portfolio

12 weeks at 8-10 hours per week · Beginner to job-ready fundamentals

Most data science roadmaps are long lists of topics with no order and no deadline. This one is a week-by-week plan: what to learn, what to build, and how to know you are ready to move on. It mirrors the structure Pathlix generates for students, so you can follow it manually or let the app schedule it around your available study hours.

What you will be able to do

  • Write clean Python for data cleaning, analysis and automation
  • Query relational data confidently with SQL joins, windows and aggregations
  • Explain and apply core statistics: distributions, sampling, hypothesis tests
  • Train, evaluate and tune supervised machine learning models
  • Ship three portfolio projects with clear write-ups and reproducible notebooks

Before you start

  • Comfort with school-level algebra
  • A laptop with Python 3 and a code editor installed
  • No prior programming experience required

The week-by-week plan

Already know an early week? Start at your first real gap instead of week one — the projects still build on each other from wherever you join.

Week 1: Python foundations

  • Variables, types and control flow
  • Functions and modules
  • Lists, dicts and comprehensions
  • Reading files and CSVs

Project: Command-line script that summarises a CSV of student marks

Week 2: NumPy and vectorised thinking

  • Arrays and dtypes
  • Broadcasting
  • Aggregations and axes
  • Random sampling

Project: Simulate 10,000 dice rolls and chart the distribution

Week 3: pandas for real data

  • Series and DataFrames
  • Indexing and filtering
  • Missing values
  • Group-by and merges

Project: Clean a messy public dataset and publish the cleaning notebook

Week 4: Exploratory data analysis

  • Descriptive statistics
  • Matplotlib and seaborn
  • Outlier detection
  • Correlation vs causation

Project: Full EDA report on a city housing dataset

Week 5: SQL for analysts

  • SELECT, WHERE, ORDER BY
  • Joins across tables
  • Aggregations and HAVING
  • Window functions

Project: Answer ten business questions against a sample e-commerce database

Week 6: Statistics that matter

  • Probability basics
  • Normal and binomial distributions
  • Confidence intervals
  • A/B testing and p-values

Project: Design and analyse a simulated A/B test

Week 7: Supervised learning: regression

  • Train/test split
  • Linear and polynomial regression
  • Loss functions
  • Overfitting and regularisation

Project: Predict house prices and document error analysis

Week 8: Supervised learning: classification

  • Logistic regression
  • Decision trees
  • Precision, recall, F1
  • Confusion matrices

Project: Build a churn classifier with a calibrated threshold

Week 9: Feature engineering and pipelines

  • Encoding categoricals
  • Scaling and imputation
  • scikit-learn pipelines
  • Cross-validation

Project: Refactor week 8's model into a reproducible pipeline

Week 10: Ensembles and tuning

  • Random forests
  • Gradient boosting
  • Hyperparameter search
  • Feature importance

Project: Enter a beginner Kaggle competition and log your score

Week 11: Communicating results

  • Storytelling with charts
  • Dashboards
  • Writing an analysis README
  • Stakeholder summaries

Project: Turn your best model into a one-page decision brief

Week 12: Portfolio and interviews

  • Cleaning up repositories
  • Case-study write-ups
  • SQL interview drills
  • Statistics interview drills

Project: Publish a portfolio site linking three finished projects

Tools you will use

PythonpandasNumPyscikit-learnPostgreSQLJupyterMatplotlibGit

Where this roadmap leads

Data Analyst

Strongest match after week 6 — SQL plus EDA covers most entry job descriptions.

Junior Data Scientist

Reachable after week 12 if all three portfolio projects are complete.

Business Intelligence Analyst

Add a dashboard tool such as Power BI or Metabase to weeks 11-12.

Frequently asked questions

How long does it take to learn data science?

With 8-10 focused hours a week, this roadmap takes about 12 weeks to reach interview-ready fundamentals. Part-time learners at 4 hours a week should plan for roughly six months.

Do I need a maths degree for data science?

No. You need comfortable school-level algebra plus the applied statistics in weeks 4 and 6. Deeper linear algebra and calculus only become important if you move into research or deep learning.

Should I learn Python or R first?

Python. It covers analysis, machine learning and production code with one language, and most Indian and global job listings ask for it.

Are certificates enough to get hired?

Rarely on their own. Hiring managers respond to three finished projects with clear write-ups far more than to course certificates, which is why every week here ends in something you build.

Want this plan scheduled for you?

Pathlix turns roadmaps like this into daily goals sized to your study hours, tracks your streak, and tests you every week.

Other roadmaps