-
CS50 Introduction to Artificial Intelligence
https://cs50.harvard.edu/ai/2024/
A high-level introduction to artificial intelligence concepts. -
Google ML Crash Course
https://developers.google.com/machine-learning/crash-course
A great introductory course to machine learning.
-
CS188 Introduction to AI
https://berkeleyai.github.io/cs188-website/
This includes graded assignments, homeworks, and projects. It requires familiarity with Data Structures and Algorithms (DSA), but you can still manage if you put effort into it. -
MIT 6.036 Introduction to Machine Learning
https://openlearninglibrary.mit.edu/courses/course-v1:MITx+6.036+1T2019/course/
This course involves deeper mathematics and the implementation of ML algorithms from scratch. -
Andrew NG ML Specialization Course
https://www.coursera.org/specializations/machine-learning-introduction
Andrew Ng explains most of the concepts with a lot of patience. It’s an awesome course for those looking to get a deeper understanding.
Once you understand the fundamentals of ML—what it is, how it works, when to use it, and why—you will likely be curious to explore further. You can go with any of the following advanced courses based on your interests.
-
MIT Micromasters in Statistics and Data Science
https://micromasters.mit.edu/ds/
This is a 5-course program spread over 12-16 months. You can start with ML or Probability, then move on to Statistics (which is a bit more difficult, but manageable with effort). There are also courses on Time Series and Data Analysis. -
DeepLearning.AI Specializations
There are multiple specializations from Deeplearning.AI. Start with the Deep Learning specialization, then explore others based on your interests, such as:- NLP (Natural Language Processing)
- Computer Vision
- Generative AI
- And more
-
CS231n: Deep Learning for Computer Vision
- Lecture Videos: https://www.youtube.com/playlist?list=PL3FW7Lu3i5JvHM8ljYj-zLfQRF3EO8sYv
- Course Website: https://cs231n.stanford.edu/
Learn about deep learning in the context of computer vision.
-
Deep Reinforcement Learning (CS285, UC Berkeley)
https://rail.eecs.berkeley.edu/deeprlcourse/
A course dedicated to deep reinforcement learning. -
Linear Algebra by Gilbert Starang, MIT 18.06
https://ocw.mit.edu/courses/18-06-linear-algebra-spring-2010/
If you're looking for jobs related to AI or ML, especially in research and solving real-world problems, a PhD degree is often required. Companies typically do not hire at the BTech level for research positions—this is a hard truth that needs to be accepted. However, for engineering roles like AI developer, ML engineer, or data scientist, you may find opportunities in startups or even in large companies. The key is to keep applying and networking.
The better path is to build a research background and pursue a PhD. You can either do a master's and then apply for a PhD, or directly apply for a PhD program. After completing a PhD, you will have multiple opportunities in large companies like Microsoft Research, Google DeepMind, and others, where you will solve real-world problems.
After doing with the courses in around year, you will feel quite comfortable and confident, then you can select a problem you want to solve, maybe it can be your personal or some common problem everyone is trying to solve. Select that problem and spend time working on it. In 1-2 years of work, you will face a lot of challenges, and there will be learning. you will realize the model training and ML is less than 10% of the actual work required to solve the problem using AI. Most of the time you will spend figuring it out, where and how to apply ML, how to convert the problem into a ML optimization problem. Mathematicall modelling and more.