Skip to content
MACHINE LEARNING · CAREER TRACK

Build models. Test what they actually learn.

Learn how data becomes a prediction, how to establish a baseline and how to evaluate a model on examples it has not seen. The result is a reproducible investigation with evidence behind its conclusions.

Enquire about this trackIs it for me?

₹30,000 per level including GST · Level 1 has no prerequisite · optional Career Audit

A portfolio you can explain

Demand prediction baseline

Estimate next-period demand and compare it with a simple historical rule.

Illustrative output: A notebook with baseline and model predictions plus an error chart.

Explore the project briefs →
THE SHORT ANSWER

What is Machine Learning as a career track?

Machine Learning is a mentor-led career track. Its focus: Train, evaluate and ship models with proper validation. First roles to search for include ML Trainee, Junior ML Associate, AI/ML Project Assistant. Level 1 takes 3 months at ₹30,000 including GST, with no prerequisite. Coding: Python practice is central to model-building work.

WHO THIS TRACK SUITS

Is Machine Learning right for you?

People who enjoy coding, patterns and investigating why a result changes, and who are willing to work with mathematics as they go.

Students
Suits CS, statistics and engineering students who already code a little. Your first project is a small model with a proper baseline and evaluation.
Freshers
Prepares you for junior data and ML-adjacent roles: you can show a notebook where the evaluation is honest and the baseline is beaten for a stated reason.
Working professionals
Fits developers and analysts moving toward modelling. Your project applies a model to data from a domain you know.
Look elsewhere if
If you want dashboards and business questions, look at Data Analytics. If you want the statistics and communication side more than the model, look at Data Science.
Coding
Python practice is central to model-building work.
Mathematics
Algebra, probability and introductory statistics.
Before you start
Practise Python variables, functions, lists and tables. Review averages, probability and graphs before tackling model evaluation. A computer capable of running small notebooks is sufficient for the illustrative projects;.
WHAT YOU WILL WORK ON

What will you learn in Machine Learning?

You learn these areas from recommended external courses; your mentor sets the practice and reviews the work.

1

Learning from data

Understand features, targets and the difference between training and evaluation.

2

Model building

Frame a prediction or classification task and train an appropriate starting model.

3

Testing and evaluation

Compare predictions with held-out outcomes and explain the chosen metric.

4

Project explanation

Document preparation, experiments, results and limitations in a model notebook.

TOOLS AND SETUP

What tools do you need for Machine Learning?

Python with the standard data and modelling libraries (pandas, scikit-learn or equivalent) in a notebook. Free to set up on your own laptop.

MENTORSHIP & REVISIONS

Bring your thinking to the review.

Check the split before the score. Feedback should examine data leakage, baseline choice, metric selection, error patterns and whether another person can reproduce the result.

See a sample review round →
LEVELS IN THIS TRACK

How long does Machine Learning take, and what does each level add?

Level 1 (3 months): small assignments and your first Machine Learning project, reviewed until it holds up. Level 2 (6 months): one larger project with revisions; resume review and mock interviews begin. Level 3 (6 months): a bigger project, portfolio guidance and application guidance for the roles you choose. Level 4 (6 months): a multi-part project with 1-on-1 reviews and system design and architecture review; aimed at working professionals switching direction. Every level is ₹30,000 including GST, bought one at a time.

What each level includes →
PRACTICE WITH PURPOSE

What projects will you build in Machine Learning?

Illustrative briefs set by mentors for practice, built on synthetic or appropriately licensed data. Your mentor sets your actual project to fit your level.

Illustrative project 01

Demand prediction baseline

Estimate next-period demand and compare it with a simple historical rule.

Sample inputs
A synthetic weekly sales series with seasonal variation and documented generation assumptions.
Your tasks
Create a chronological split, build a naive forecast, train a small regression model and compare errors.
What to deliver
A notebook with baseline and model predictions plus an error chart.
Quality checks
Use only information available at prediction time; report MAE and errors during unusual periods.
Interview discussion
Explain why a random split could misrepresent future performance.
Illustrative project 02

Subscription churn study

Identify which fictional accounts may stop using a service.

Sample inputs
Synthetic usage records with a clearly defined observation window and later churn label.
Your tasks
Define the target, remove post-outcome fields, compare a simple classifier with a majority baseline and examine thresholds.
What to deliver
A model report with a confusion matrix and threshold discussion.
Quality checks
Check class balance, precision, recall and leakage from cancellation records.
Interview discussion
Explain the tradeoff between missed churn and unnecessary retention outreach.
Illustrative project 03

Customer grouping experiment

Explore whether purchase patterns form useful descriptive groups.

Sample inputs
Synthetic customer-level frequency and spending features without personal identifiers.
Your tasks
Scale features, explore clustering, describe each group and compare stability across choices.
What to deliver
An exploratory notebook and a cautious segment summary.
Quality checks
Check sensitivity to scale and outliers; do not treat clusters as ground-truth customer types.
Interview discussion
Explain why a mathematically separated cluster may not be useful to a business.
FIRST ROLES, HONESTLY

What jobs can Machine Learning lead to?

Machine Learning Engineer roles in India typically expect prior experience. The track prepares you for junior data and ML-adjacent roles first, where a correctly evaluated model and a clear write-up are what get noticed.

Preparing datasets, running experiments and documenting evaluations under supervision. ML trainee and AI/ML project assistant are useful search terms; production ML engineering usually requires broader software and deployment skills.

We do not guarantee eligibility, interviews or employment.

YOUR PORTFOLIO

Make your contribution visible.

A reproducible notebook, data description, baseline comparison, evaluation report and a README with clear setup instructions.

Career preparation by levelAll project briefs
PRACTISE THE CONVERSATION

What you will be asked about this work.

The questions attached to each project brief are the ones a mentor asks in review and an interviewer asks later. From Level 2, you rehearse them in mock interviews.

  1. 1

    ABOUT “DEMAND PREDICTION BASELINE”

    Explain why a random split could misrepresent future performance.

  2. 2

    ABOUT “SUBSCRIPTION CHURN STUDY”

    Explain the tradeoff between missed churn and unnecessary retention outreach.

  3. 3

    ABOUT “CUSTOMER GROUPING EXPERIMENT”

    Explain why a mathematically separated cluster may not be useful to a business.

QUESTIONS ABOUT THIS TRACK

Before you choose Machine Learning.

Must I know Python before joining?

Python practice will help. Tell the team your current preparation on the call; Level 1 has no prerequisite and the mentor sets the first project to match it.

How much mathematics should I prepare?

Start with algebra, probability, averages and variation. Develop the mathematics alongside model examples.

Why do I need a baseline?

It tells you whether a model improves on a simple rule. Complexity alone is not evidence of value.

Is high accuracy always good?

No. A majority-class prediction can look accurate on imbalanced data. Inspect the errors and appropriate metrics.

Can I train on all the data?

Keep evaluation data separate during development. Otherwise you cannot estimate how the model behaves on unseen examples.

Do I need an expensive GPU?

Small tabular learning projects can use a normal computer. Confirm hardware needs for the actual programme.

Does this qualify me immediately as an ML engineer?

No single programme establishes that. Engineering roles also assess coding, systems, deployment and relevant experience.

What should I put in the repository?

Include setup, data provenance, split rules, experiments, metrics, limitations and instructions to reproduce results.

KEEP EXPLORING

Guides to read alongside Machine Learning

BY CITY

Where is Machine Learning a good fit?

These cities have employers that hire for the roles this track leads to.

Take your next step in Machine Learning.

Enquire about this track. The team calls you back, usually within one working day, to confirm the starting level.

Enquire about this track
Enquire now