AI career path · Updated
How to start an AI career from scratch in 2026
Build the base, then land a junior AI / ML role.
Quick answer
AI Foundations Path: build the base, then land a junior AI / ML role. To start an AI career from scratch, follow 8 steps: Python Basics; Git, GitHub & Data Handling; Statistics & Probability Basics; SQL & Data Basics; Machine Learning Fundamentals; Intro to LLMs & Generative AI; Build 2 Portfolio Projects; Apply for Junior Roles. Expect 4.8 months at 10 hours a week, depending on your starting skills. Best for: students, graduates and people new to coding.
Key facts
| Best for | Students, graduates and people new to coding |
|---|---|
| Time to job-ready | 4.8 months at 10 h/week |
| Coding needed | Basic Python |
| Maths needed | Basic statistics |
| Demand | Competitive — projects make the difference |
| Pay (US) | US junior roles: roughly $80k–$130k base |
| Pay (India) | Freshers ₹4–10 LPA; ₹8–15 LPA with strong GenAI projects at product companies |
| Job titles | Junior ML Engineer, AI Engineer Trainee, Data Analyst (AI), ML Intern |
Pay is indicative, compiled from Levels.fyi, Glassdoor, AmbitionBox and 2026 salary guides, which often disagree. It varies a lot by city, company type and experience.
What does a junior AI / ML engineer do?
You are early in your tech journey, so the fastest route is a solid foundation: Python, data, and core machine learning, proven with portfolio projects. This path leads to Junior ML Engineer, AI Engineer Trainee or Data Analyst (AI) roles.
Core skills
- Python
- Statistics
- Machine learning basics
- SQL
- Git & GitHub
Who usually moves into this role
Not sure this is the one? Read how to choose the right AI role.
AI Foundations Path roadmap: 8 steps
Follow the steps in order. Times assume about 10 study hours a week. Take the free questionnaire to get a personalised version that skips what you already know and fits your weekly hours.
-
Step 1: Python Basics 3-4 weeks
Variables, loops, functions, data structures. Python is the #1 language in AI because most ML libraries are written in it.
Why it matters: Every later step is written in Python — without it nothing else sticks.
Learn from: CS50P Harvard (free), freeCodeCamp Python (free)
Topics: Variables & data types, Control flow (if/else, loops), Functions & parameters, Lists, dictionaries, sets, File I/O, Basic error handling
Must-know: Write clean, readable code; Finish 2 mini projects (e.g. expense tracker)
-
Step 2: Git, GitHub & Data Handling 2 weeks
Version control and working with data in Pandas and NumPy.
Why it matters: Employers look at your GitHub first, and Pandas is how all AI work touches data.
Learn from: GitHub Skills (free), Kaggle Learn Pandas (free)
Topics: git add / commit / push, Pull requests, Pandas DataFrames, Cleaning missing values
Must-know: Public GitHub profile; Clean a real CSV dataset
-
Step 3: Statistics & Probability Basics 2 weeks
Mean, median, variance, distributions, sampling. Statistics tells you whether a model's result is real or random noise.
Why it matters: Without it you cannot tell a good model from a lucky one — interviewers test this.
Learn from: Khan Academy Statistics (free), StatQuest YouTube (free)
Topics: Mean, median, mode, Variance & standard deviation, Probability distributions, Sampling & bias, Correlation vs causation
Must-know: Know when to use mean vs median; Explain correlation vs causation
-
Step 4: SQL & Data Basics 2 weeks
Querying data from databases. AI needs data, and most company data lives in SQL databases.
Why it matters: SQL appears in nearly every junior data/AI interview and job description.
Learn from: SQLBolt (free), Mode SQL tutorial (free)
Topics: SELECT, WHERE, GROUP BY, JOINs, Aggregations, Subqueries
Must-know: Master JOINs and GROUP BY
-
Step 5: Machine Learning Fundamentals 5-6 weeks
Supervised learning: regression and classification, train/test split, overfitting, evaluation metrics, using scikit-learn.
Why it matters: This is the core of every AI role and what separates you from people who only use ChatGPT.
Learn from: Andrew Ng Machine Learning Specialization (audit free), Google ML Crash Course (free)
Topics: Supervised vs unsupervised, Linear & logistic regression, Decision trees, Train / test split, Overfitting, Precision, recall, F1
Must-know: Accuracy alone is not enough; Use scikit-learn pipelines
-
Step 6: Intro to LLMs & Generative AI 2 weeks
How large language models work at a high level, prompting, and calling an LLM API from Python.
Why it matters: Most new AI job posts mention LLMs — even junior candidates are expected to have tried them.
Learn from: DeepLearning.AI short courses (free), Anthropic / OpenAI API docs (free)
Topics: Tokens & context windows, Prompting patterns, Calling an API from Python, Basic RAG idea
Must-know: Ship one tiny LLM app
-
Step 7: Build 2 Portfolio Projects 4 weeks
One classic ML project and one LLM project, each with a clear README and a live demo.
Why it matters: Projects are your proof of skill when you do not have AI work experience.
Learn from: Kaggle datasets (free), Streamlit / Hugging Face Spaces (free hosting)
Topics: Problem statement, Data cleaning, Model & evaluation, README with results, Deploy a demo
Must-know: Live demo link on your resume
Project ideasLoan default predictor: Classic ML on a public lending dataset with a Streamlit demoPDF question-answering bot: Small RAG app over your college notes or a public report -
Step 8: Apply for Junior Roles ongoing
Target: Junior ML Engineer, AI Engineer Trainee, ML Intern, Data Analyst (AI track).
Why it matters: Applying while you keep building compounds — interviews show you what to learn next.
Learn from: LinkedIn, Naukri, Wellfound, company career pages
Topics: Resume with project results, LinkedIn posts about what you built, Interview practice
Must-know: Projects at the top of your resume
Frequently asked questions
What does a junior AI / ML engineer do?
You are early in your tech journey, so the fastest route is a solid foundation: Python, data, and core machine learning, proven with portfolio projects. This path leads to Junior ML Engineer, AI Engineer Trainee or Data Analyst (AI) roles.
How long does it take to start an AI career from scratch?
4.8 months at 10 hours of study a week. The longer estimate is for starting from zero. The shorter one applies if you already have the core skills (basic Python and the listed basics), because those steps become quick revision.
Do I need coding and maths to start an AI career from scratch?
Coding: Basic Python. Maths: Basic statistics.
How much does a junior AI / ML engineer earn?
Indicative pay in the US: US junior roles: roughly $80k–$130k base. In India: Freshers ₹4–10 LPA; ₹8–15 LPA with strong GenAI projects at product companies. Pay varies widely by city, company type and experience.
Which job titles should I search for?
Junior ML Engineer, AI Engineer Trainee, Data Analyst (AI), ML Intern.
Who is the AI Foundations path best for?
Students, graduates and people new to coding. Build the base, then land a junior AI / ML role.
Is AI Foundations Path the right AI role for you?
Our free AI career consultation asks about 20 questions on your experience, skills, working style and goals. You get your best-fit AI role out of 9, the reasons behind it, your skill gaps, and a dated roadmap. It takes about 4 minutes and needs no sign-up.
Start the free consultation