Getting Started with AI: Tips from a Fellow Learner

Hi everyone,

I've noticed many people in the community seem a bit lost on how to get started with AI, so I decided to write this post in the hope that it might help.

There are two types of people I see in this space: those who want to know how to use AI (like using ChatGPT, for example) and those who want to work with AI, meaning they want to create models, develop AI applications, and understand the underlying technology. These are different goals, so it's important to be clear about what you're aiming for first.

If you're more interested in how to use AI, such as with tools like ChatGPT, I suggest checking out this resource: Learn Prompting Basics.

For those who want to work with AI, whether by building applications or developing models. I recommend starting with an area that interests you. You can always shift focus later once you explore more options.

This post also serves as a kind of response to this other one: How to Get Started with AI?

Here's my take on a structured yet flexible approach:

Start with Python:

Python is the primary language for AI. Get comfortable with its basics and libraries like:

NumPy and Pandas for data manipulation.
Matplotlib and Seaborn for data visualization.

Learn Key Libraries for AI/ML After mastering Python, focus on:

Scikit-learn for traditional machine learning algorithms.
TensorFlow or PyTorch for deep learning frameworks.

Refresh Your Math Skills:

Linear Algebra for neural networks.
Calculus for optimization techniques.
Probability and Statistics for model evaluation.

Explore Machine Learning:

Key Algorithms: Familiarize yourself with foundational algorithms, including:

Linear Regression: Used for predicting continuous outcomes.
Decision Trees: Useful for classification and regression tasks.
Support Vector Machines (SVM): Effective for high-dimensional spaces and classification tasks.
Clustering Algorithms: Such as K-Means and Hierarchical clustering for grouping data points.

Model Evaluation: Understand how to evaluate your models using techniques like cross-validation, confusion matrices, and performance metrics (e.g., accuracy, precision, recall).

Feature Engineering: Learn how to prepare your data for machine learning, including techniques like normalization, encoding categorical variables, and creating interaction features.

Explore more...

Focus on Practice and Use AI to Guide Your Learning:

I believe that hands-on practice is the best way to learn. Applying the theory through projects/exercises makes learning more enjoyable and effective.

AI itself can be a fantastic learning tool. You can ask AI (like ChatGPT) to help you design a learning path tailored to your goals. Whether it's Computer Vision, Natural Language Processing (NLP), or reinforcement learning, you can adjust your approach to what works best for you.

For example, two weeks ago, I decided to dive into Computer Vision. I asked ChatGPT to help me create a practical, exercise-based roadmap, and now I’m learning by building small projects alongside reading books and theory. I believe that mistakes and corrections are where the real learning happens.

Have Clear Goals and Focus on What You Want to Achieve:

If you don’t have a goal, it’s easy to feel lost. Whether you want to build a specific type of AI application, solve a particular problem, or just understand the technology, define your goals early. Without direction, learning can become overwhelming, so set clear objectives and work toward them step by step.

I hope this helps! I'm not an expert whatsoever, just someone learning like you. So, don't take my words as the final answer. Explore what works for you and, most importantly, take action.

What has been your experience with learning AI or machine learning? I’d love to hear about your journey or any tips you have!

Have a great week!

17 comments