Skip to content
New

Python for Data Science

Pandas, statistics and models that ship

4.0(1)Rated 4.0 out of 5 from 1 reviews2 learners12 lessons · 3m
Created by Demo TrainerUpdated August 2026EnglishBeginner

What you'll learn

  • Clean and reshape data that arrives broken
  • Choose a model for the problem rather than the fashion
  • Evaluate honestly, including on data the model has not seen
  • Serve a trained model behind a monitored API
Level
Beginner
Duration
14 weeks
Material
3m
Lessons
12
Language
English
Access
365 days

About this course

Work with real, messy data: clean it, understand it, model it, and put the model behind an endpoint somebody else can call.

Most data courses stop at the notebook. This one does not. You will start with genuinely messy data, learn the statistics that stop you fooling yourself, train models that hold up on data they have never seen, and finish by serving one behind an API with monitoring on it.

Syllabus

4 modules · 12 lessons · 3m · 1 free to preview

  1. Getting set up1 free3 lessons · 3m
  2. The core ideas3 lessons
    • What you are going to build
    • Setting up your machine
    • Your first working version
  3. Building it for real3 lessons
    • What you are going to build
    • Setting up your machine
    • Your first working version
  4. Shipping and what comes after3 lessons
    • What you are going to build
    • Setting up your machine
    • Your first working version

Taught by

  • Gs DwiwediLead

    Full-stack engineer, and the person who built this platform

    15 years experience · Java, Spring Boot, React, PostgreSQL, System design

    Fifteen years building and running production systems, most of them in Java and JavaScript. Teaches the way he works: start with something that runs, then make it correct, then make it fast.

  • Meera Iyer

    Data scientist — models that reach production

    8 years experience · Python, pandas, scikit-learn, MLOps, Statistics

    Spent eight years shipping models that other teams depend on, which is a very different job from winning a leaderboard. Teaches evaluation before algorithms.

What you'll get

  • Lifetime accessCome back to the material whenever you need it.
  • Reviewed projectsA person reads your code and tells you what to change.
  • CertificateIssued on completion, verifiable by a link.
  • Doubt sessionsLive weekly, recorded if you cannot make it.

Tools you'll use

  • PythonThe language, properly — not just notebook snippets
  • pandasReshaping and cleaning data at scale
  • scikit-learnModels, pipelines and honest evaluation
  • FastAPIServing a model as an endpoint

Before you start

  • Basic programming in any language
  • School-level mathematics

Who this is for

  • Analysts moving into modelling
  • Developers adding data work to their range