Amazon ML Summer School 2026: The Complete Guide to Apply, Prepare & Get Selected

Lets Code•June 16, 2026

If you’re an engineering student in India who is even a little serious about machine learning, Amazon ML Summer School (MLSS) is one of the few programs genuinely worth chasing. You get trained by Amazon’s own Applied Scientists, it’s completely free, and a strong showing here adds real weight to your profile when you start applying for internships and full-time roles.

This is the only guide you’ll need. We’ll cover what the program actually is, who can apply, the exact selection process and test pattern, the full syllabus, a realistic preparation plan, and a curated list of free resources — including the Let’s Code tools that make each stage easier.

ℹ️ Where the 2026 cycle stands right now: Registration for the 2026 edition closed on 14 June 2026 and the SOP round ran the same evening. The Selection Test window is 28 June – 28 July 2026. If you registered, your focus now should be 100% on test prep (jump to the preparation section). If you missed this cycle, bookmark this guide — the program runs annually, and everything below applies to the next edition too.

Amazon ML Summer School at a Glance

Detail	Information
Program	Amazon ML Summer School (MLSS) 2026 — 6th edition
Conducted by	Amazon India (Applied Scientists & ML researchers)
Mode	Fully online / virtual
Cost	100% free
Who can apply	B.Tech/B.E, M.Tech/M.E/MS, or PhD students at any recognized Indian institute
Graduation year	Expected to graduate in 2027 or 2028
Selection	3 stages — Resume → SOP (500 words) → Selection Test (60 min)
Seats	~3,000 (historically), highly competitive
You get	Live expert-led modules, completion certificate, acknowledgement letter, Amazon swag, networking with scientists

What Is Amazon ML Summer School?

Amazon launched the program in 2021 to give Indian engineering students a structured, no-cost path into machine learning taught by people who build ML systems at scale. It’s an intensive, expert-led course on core ML topics, delivered through live sessions, real Amazon case studies, and curated reading.

The key thing to understand: MLSS is a learning program, not a recruitment drive. Selection does not guarantee an Amazon internship. But what it does do is build genuine ML fundamentals, put a recognized line on your resume, and connect you to scientists and a peer network — all of which compound when you start interviewing. Sessions are taught around real systems most people have used, like Alexa, Amazon Go, and AWS recommendation engines, so the theory always lands on something concrete.

If your college courses feel disconnected from what the industry actually expects, this program is built to close exactly that gap.

Who Can Apply? (Eligibility)

For the 2026 edition, you are eligible if you are:

Enrolled in a Bachelor’s, Master’s, or PhD program (B.Tech/B.E, M.Tech/M.E/MS, PhD)
Studying at any recognized institute in India
Expected to graduate in 2027 or 2028

A few clarifications that trip people up:

Any branch is welcome. You don’t have to be CSE. Students from ECE, EE, Mechanical, and other disciplines apply and get in.
No prior ML experience is required to apply. Familiarity with Python and basic math helps a lot during both the test and the sessions, but it isn’t a hard gate.
Eligibility years shift each edition. Always confirm the current graduating-year requirement on the official Unstop listing before applying, since it moves forward every year.

💡 Tip: Not sure if your profile is competitive yet? Run it through the Job Ready Score tool on Let’s Code to see where your fundamentals, projects, and resume stand — and what to fix before you apply.

Important Dates (2026 Edition)

Stage	Window
Stage 1 — Registration	1 June 2026 → 14 June 2026, 12:00 PM IST
Stage 2 — SOP Submission	14 June 2026, 4:00 PM – 8:00 PM IST
Stage 3 — Selection Test	28 June 2026, 12:00 PM IST → 28 July 2026, 5:00 PM IST

⚠️ Deadlines are firm. Amazon does not extend MLSS deadlines and late applications are not entertained. Results aren’t published publicly — all shortlist communication comes to your registered email and the Unstop platform after each stage. Check both daily during the cycle.

The Selection Process (3 Stages)

MLSS uses a funnel. You must clear each stage to advance to the next.

Stage 1 — Resume Submission

You submit a digital-friendly resume on Unstop. Shortlisting at this stage is based on your resume, so Amazon is reading your academic background, relevant coursework, projects, technical skills, and any prior exposure to ML or programming. Keep it clean, ATS-friendly, and technically focused. Drop the fluff; lead with projects and skills.

Use Resume Studio on Let’s Code to build an ATS-friendly, single-page technical resume in minutes — it’s tuned for exactly this kind of screening.

Stage 2 — Statement of Purpose (SOP)

Registered candidates submit a 500-word SOP in PDF format. A representative prompt asks you to describe your technical journey in AI/ML so far: what you’ve built or explored, what gaps remain in your knowledge that MLSS will help close, and why your curiosity and experience make you a strong fit.

This is where most generic applications die. See the SOP tips below.

Stage 3 — Selection Test

Shortlisted candidates take a 60-minute online assessment (typically on platforms like HackerRank or Mercer–Mettl). This is the make-or-break round. Full pattern next.

Test Pattern: Exactly What to Expect

The selection test is 60 minutes and split into two parts:

Part A — ~20 MCQs on basic ML concepts, probability, statistics, and linear algebra
Part B — Coding / programming questions (typically 1–2 problems testing DSA and problem-solving)

The single most important insight, drawn from past candidates’ experiences: students who only memorize ML theory tend to struggle, while those with strong fundamentals in probability, statistics, linear algebra, and DSA consistently do well. Amazon tests application of concepts, not textbook definitions. Time is tight, so speed and accuracy both matter.

💡 Tip: Want to feel the pressure before test day? Practice timed problem-solving and ML conceptual questions with the Mock Interview tool, and review past-pattern questions in the PYQs section on Let’s Code.

Syllabus: What You’ll Learn

Once you’re in, the curriculum runs across roughly eight modules, taught with theory + hands-on demos + Amazon’s real ML case studies. Expect to cover:

Supervised Learning — regression, classification, model evaluation
Deep Neural Networks — architectures, training, backpropagation
Dimensionality Reduction — PCA and friends
Unsupervised Learning — clustering and density estimation
Probabilistic Graphical Models
Sequential Models — RNNs, attention, the building blocks behind modern NLP
Reinforcement Learning
Generative AI — the area everyone wants to understand right now

The same topics that appear in the syllabus overlap heavily with what’s tested in the selection round — so the prep below does double duty.

How to Prepare: A Realistic Plan

You don’t need to be an ML expert. You need solid fundamentals + sharp problem-solving. Here’s where to spend your time, in priority order.

1. Math foundations (highest ROI)

This is what most candidates underrate.

Linear Algebra — vectors, matrices, eigenvalues/eigenvectors, matrix operations, projections.
Probability — conditional probability, Bayes’ theorem, distributions, expectation/variance.
Statistics — descriptive stats, hypothesis testing basics, estimators, correlation vs. causation.

2. Core ML concepts

Difference between supervised, unsupervised, and reinforcement learning
Bias–variance tradeoff, overfitting/underfitting, regularization
Common algorithms: linear/logistic regression, decision trees, k-NN, k-means, basic neural nets
Evaluation metrics: accuracy, precision/recall, F1, ROC-AUC, confusion matrix

3. DSA & coding

Strong DSA is frequently the difference between selection and rejection.

Arrays, strings, hashing, two pointers
Sorting & searching
Recursion, basic DP
Time/space complexity analysis
Be fast and clean in Python (the most common choice for this test)

4. A sensible 3–4 week schedule

If you have a month before the test:

Week 1: Linear algebra + probability (concepts + practice problems daily)
Week 2: Statistics + core ML concepts (revise one algorithm family per day)
Week 3: DSA grind — 3–5 problems/day, focus on patterns not volume
Week 4: Full-length timed mocks, mixed MCQs + coding, review weak spots

ℹ️ Note: Track your prep deadlines and the test window using the Internship/Deadline Tracker on Let’s Code so a tight 4-hour SOP window or a test date never slips past you.

Nailing the SOP

500 words is short. Make every sentence earn its place.

Be specific, not generic. “I am passionate about AI” says nothing. “I built a sentiment classifier on 10k product reviews and got stuck on class imbalance” says everything.
Show a journey + a gap. What have you done, what are you missing, and how does MLSS fill that gap? The prompt literally asks for this — answer it directly.
Tie your interest to something real. A project, a paper you struggled through, a problem you couldn’t solve. Concrete beats aspirational.
Proofread ruthlessly. Submit clean PDF, under 500 words, no typos.

The Cover Letter AI tool on Let’s Code can help you draft and tighten a first version of your SOP — then make it personal in your own voice.

Free Resources (Curated)

You can prepare for the entire test using free material. Here’s the shortlist worth your time.

Math for ML

Mathematics for Machine Learning (Deisenroth, Faisal, Ong) — free PDF, the single best math reference for this
3Blue1Brown — “Essence of Linear Algebra” (YouTube) — intuition you won’t forget
Khan Academy — probability & statistics tracks
StatQuest with Josh Starmer (YouTube) — the friendliest stats + ML explanations anywhere
Imperial College London — Mathematics for Machine Learning (Coursera, audit for free) — linear algebra, multivariate calculus, and PCA in one track

Machine Learning

Andrew Ng’s Machine Learning Specialization (Coursera / DeepLearning.AI) — the canonical starting point
An Introduction to Statistical Learning (ISLR) — free PDF, excellent for concepts
CS229 (Stanford) lecture notes — free, for going deeper
Google Machine Learning Crash Course — fast, interactive, and free

Deep Learning & Modern Topics

DeepLearning.AI Deep Learning Specialization (audit for free)
fast.ai — Practical Deep Learning — hands-on and free

Coding / DSA

Kaggle’s free Python micro-course + practice notebooks
NeetCode — DSA patterns & roadmap — pattern-first practice, exactly what this test rewards
Standard DSA practice on your platform of choice — focus on patterns

Amazon’s own material

Amazon Science blog — visual essays explaining ML concepts, plus context on how Amazon uses ML

Let’s Code (for the application stages)

Resume Studio — ATS-friendly resume for Stage 1
Cover Letter AI — draft your SOP for Stage 2
Mock Interview + PYQs — drill for the Stage 3 test
Job Ready Score — gut-check your overall readiness

Likely Question Patterns (PYQ-Style)

Based on past patterns, expect MCQs and coding problems shaped like these (illustrative — not actual past questions):

Given a confusion matrix, compute precision, recall, and F1.
A bag-of-balls / dice setup testing conditional probability and Bayes’ theorem.
“Which of these reduces overfitting?” — regularization, dropout, more data, etc.
Eigenvalue / matrix-rank conceptual MCQ.
A medium-difficulty array or string coding problem with a time-complexity constraint.

The pattern is clear: applied probability + linear algebra MCQs, plus a couple of clean DSA problems. Prepare for application, not recall.

What You Get Out of It

A spot in a genuinely respected, scientist-led ML program — for free
A completion certificate and an official acknowledgement letter from Amazon
Exclusive Amazon swag on successful completion
Networking with Amazon scientists and high-performing peers
Real ML fundamentals that make you visibly stronger in internship and placement interviews

It won’t hand you an internship — but it’s one of the cleanest ways to level up your profile and your actual skills at the same time.

Common Mistakes to Avoid

Registering at the last hour. Unstop slows down near deadlines and the window is firm. Don’t risk it.
Submitting a generic resume/SOP. Both are real filters. Tailor them.
Studying only ML theory. The test rewards math + DSA. Don’t skip them.
Ignoring your registered email. Every shortlist notification goes there. Check daily.
Treating it as a guaranteed internship. Go in to learn; the profile boost follows.

FAQ

Is Amazon ML Summer School free? Yes — completely free for selected participants.

Does it guarantee an Amazon internship? No. It’s a learning program. It strengthens your ML knowledge and profile, but selection ≠ internship.

Do I need prior ML experience to apply? No. But strong fundamentals in math, Python, and DSA help a lot during the test and the sessions.

Which branches can apply? Any engineering discipline at a recognized Indian institute, as long as you meet the graduation-year criteria.

Are there coding questions in the test? Yes — the selection test typically includes 1–2 programming problems alongside the MCQs, so DSA prep matters.

When are results announced? Results aren’t public. Shortlisted candidates are notified via registered email and the Unstop platform after each stage.

Final Words

Amazon ML Summer School is one of those rare programs where the prize is real learning from people who actually build ML at scale — not just another certificate. The competition is steep (~3,000 seats from a huge pool), but the bar is fundamentals, not genius. Get your math, ML concepts, and DSA solid, write a specific SOP, submit a clean resume, and you give yourself a genuine shot.

If you’re starting your prep now, build your edge with the free tools on Let’s Code — from Resume Studio and Mock Interview practice to your Job Ready Score. And when applications open for the next edition, you’ll be the one who’s ready before the deadline panic begins.

Good luck — go build something worth writing about.

Join our WhatsApp channel for more resources – Link

Lets Code

Contributing Writer