How to Effectively Learn Data Science in 2024

Self-Study Guide to Get Ready for Data Science Jobs

Richard Warepam
9 min readDec 30, 2023


Image by Author

I’ve always been someone who learns on their own. Everything I know about programming and business, I picked up without a tutor or mentor. Sure, it wasn’t easy, but I loved every step of my self-learning journey.

How about you? Are you also walking the self-learner path?

Let me share something exciting with you. Self-learning today is a breeze compared to how it was back in my early days. And the game changer? — AI tools, especially ChatGPT.

It’s like having a helper, mentor, and teacher all rolled into one. So, why not leverage ChatGPT to boost our learning journeys in 2024?

In this article, I’m going to dive into “what topics to start with” and “how to approach them,” plus I’ll throw in some super helpful tips.

Ready to embark on this journey? Let’s dive in!

Understanding the Basics

Here, I won’t be explaining all the topics; rather, I will point out the things you need to learn to get started with your data science journey.

1. Statistics:

Now, let’s talk about a big oops moment many self-learners have when they dive into data science.

Often, they skip over “statistics” and head straight for Python, SQL, or other tech stuff.

But here’s my advice: Begin with “Statistics.”

Really spend time on it and get it down pat before you leap into other areas. It’s a crucial first step to becoming an awesome data wizard.

Curious about why this matters so much? Check out my article here:

🎯 To-do Checklist for Statistics:
a. Probability theory
b. Descriptive Statistics
c. Inferential Statistics
d. Statistical Machine Learning

📚 Resource: Interview Ready Notes: Practical Statistics for Data Scientists

2. Programming Skills

So, you’ve got the hang of statistics? Great! Next up, it’s time to pick a programming language to boost your data science skills.

You’ve got two really good choices here: Python and R.

Now, which one should you go for? It’s totally up to what you feel comfortable with.

But, if you ask me, I’d say Python is a super choice.

Why? — Because it’s super versatile and easy to get the hang of. Plus, Python comes packed with loads of libraries. These are like toolkits that make your work a lot easier, whether you’re handling data, doing complex calculations, or anything else.

On the other hand, there’s R. R is fantastic, especially if you’re all about statistics. It’s like the go-to language for statistical analysis. But remember, it’s mostly just about statistics.

So, think about what you need and pick the one that feels right for you.

If you’re leaning towards being a jack-of-all-trades in data science, Python might just be your best bet!

🎯 To-do Checklist for Python:
a. Python Fundamentals
b. Pandas and NumPy Library (DataFrame Basics and operations)
c. Visualization (Matplotlib and Seaborn Libraries)
d. Data Scraping (BeautifulSoup, Scrapy, Selenium or Requests libraries)
e. Error Handling and Debugging

📚 Resource: 50 Days of Data Analysis with Python: The Ultimate Challenge Book for Beginners

3. EDA — Data Wrangling and Visualisation

You’ve just taken your first steps into the world of Pythoncongratulations! Now, let’s talk about what comes next.

As someone new to data science, it’s easy to think it’s all about diving into data to find those eye-opening insights.

You might be tempted to spend most of your time analyzing data or crafting complex models.

But there’s something even more fundamental to learn first: EDA, or exploratory data analysis.

EDA is the backbone of data science work in every company.

It involves cleaning, summarizing, transforming, and visualizing data.

These tasks might not sound as glamorous as building models, but they are crucial.

In fact, for beginners like you, mastering EDA is a key step to landing your first job in the field.

While analyzing trends and building models are part of data science, they often fall to more experienced professionals. So, focus on becoming great at EDAit’s your ticket to a successful start in data science!

If you want to learn how to conduct an effective EDA, read this article of mine:

📝 My Articles:
How to Conduct an Effective Exploratory Data Analysis (EDA)
How to Visualize Data in the Most Effective Way

🎯 To-do Checklist for EDA:
a. Data Summarization
b. Data Cleaning
c. Data Transformation
d. Data Visualization

📚 Resource: 50 Days of Data Analysis with Python: The Ultimate Challenge Book for Beginners

4. SQL (Data Manipulation and Extraction)

In addition to Python, there’s another key player in the programming world: SQL (Structured Query Language).

If you’ve mastered SQL, guess what? You’ve just unlocked a treasure trove of job opportunities!

SQL is a skill that’s in hot demand across all industries. It’s the go-to tool for for querying and manipulating databases.

Being able to read, write, and optimize SQL queries is crucial for pulling out and tweaking data.

It’s a skill that really boosts your data game!

🎯 To-do Checklist for SQL:
a. Big 6 Statements: (SELECT, FROM, WHERE, GROUP BY, HAVING, and ORDER BY)
b. Joins and CTEs
c. Window Functions
d. Stored Procedures

📚 Resources: Minimum Viable SQL Patterns

📝 My Articles:
The Ultimate Guide to Mastering “CASE WHEN” in SQL for Data Wizards
Mastering SQL “WINDOW FUNCTIONS”: The Ultimate Guide

Alright, if you’ve mastered all the skills we’ve talked about, you’re on track to become a “data analyst”.

Just a heads up, though — don’t forget to learn a visualization or Report generating tool, like PowerBI or Tableau. They’re crucial!

But, aiming to be a “data scientist” or “data engineer”? That’s a different ball game. You’ll need some extra, more advanced skills. Let’s dive into that now.

Advanced Skills

From here, the learning path becomes more difficult and complex because these topics are not very suitable for a beginner.

For these skills, one needs to have a better understanding of mathematical topics like linear algebra, calculus, and even some prior computation theory knowledge. Let’s dive in.

1. Machine Learning

Now that you’ve mastered the basics, you’re all set with skills like data scraping, cleaning, and statistical analysis.

You know the drill: turning raw data into something useful.

The next big step?

Using this data to build models that unlock deeper insights and drive smart business choices.

This is where “machine learning” comes into play. It’s all about teaching computers to think and learn from data, just like humans.

The journey involves understanding various algorithms, from simple ones like linear regression to complex neural networks (that’s deep learning for you).

Sure, these concepts might seem tough, but they are cutting-edge technology.

Embrace the challenge and learn, or risk falling behind. The choice is yours!

🎯 To-do Checklist for Machine Learning:
a. Feature Engineering
b. Supervised and Unsupervised learning
c. Regression algorithms (Linear Regression, Logistic Regression, etc.)
d. Classification algorithms (Logistic Classification, SVM, Naive Bayes, etc.)
e. Clustering algorithms ( K-means, mainly)
f. Deep Learning Concepts ( ANNs, CNNs, RNNs, Transformers, PyTorch/Tensor-flow Basics)

📚 Resources: I am working on it. It will be available here in 2 months (Starting in 2024)Keep it Subscribed, to get notifications.

2. Model Evaluation

Once you’ve created your machine learning models, it’s natural to wonder how well they’re doing.

It’s tricky because what goes on inside these models can be pretty complex.

That’s where the importance of evaluating your models comes in.

In data science, making sure your models are doing their job right is crucial. This means you should definitely get to know the ‘model evaluation methods’.

They’re key to understanding and improving your models!

The information you need to understand about it is:

  1. Which evaluation methods are best to use for different situations,”
  2. How to evaluate the models,” and lastly,
  3. How to interpret these evaluations.

This information will guide you in improving your model to achieve your desired goal.

Learn the basics from my article:
How to Select the Best Model Evaluating Methods and When to Use Them: The Ultimate Guide

🎯 To-do Checklist for Model Evaluation:
a. Confusion Matrix
b. Precision, Recall, and F1-Score
c. Cross-validation
d. Overfitting, Underfitting

📚 Resources: I am working on it. It will be available here in 2 months (Starting in 2024)Keep it subscribed to get notifications.

Now, let’s talk about two advanced topics for those dreaming of becoming data scientists.

If you’ve made it this far, you’re nearly set to begin your journey as a data scientist.

But wait, there’s one more thing. Beyond the basics, there’s an advanced topic that’s key for data engineers. And that is:

3. Big Data Technologies:

Let’s dive into the role of a data engineer.

Their main job?

Handling the ‘engineering’ side of data. This involves tasks like gathering data from various sources and setting up automated processes.

Essentially, they build a data flow or pipeline to collect all this data in one place. That’s where learning about ‘big data technologies’ becomes crucial.

Why ‘big data’, you ask?

Well, today’s world is overflowing with data, and it’s in massive quantities — that’s why it’s called ‘Big Data’.

To manage this, you’ve got to get familiar with several technologies. I know it sounds like a lot to take in.

But here’s a piece of friendly advice: When you’re learning these technologies, concentrate on understanding their fundamental concepts.

These foundational concepts stay the same, even though the technologies themselves are always evolving and changing.

This approach will give you a solid base to adapt and grow with the technology.

Learn more about it in my article:
How to Break into Data Engineering: 2024

🎯 To-do Checklist for Big Data Technologies:
a. Big Data Introduction
b. Distributed System
c. Hadoop (Map Reduce)
d. Spark
e. Cloud Computing

📚 Resource: How to Break into Data Engineering: 2024

👉 Note:

Apart from the prompts given below, if you want the “collection of Questions” of each topic that you can use while learning with ChatGPT. Checkout my eBook:

Contents of the eBook

📚 Get the eBook now: ChatGPT for Learning Data Science

New Year Offer till 15th Jan. 2024:

Get 50% OFF
Use Code: “NEWYEAR50”

Customizable ChatGPT Prompts:

1. Clarifying Concepts

“Please explain the concept of [Topic] in a way that is understandable for someone with [Basic/Intermediate/Advanced] knowledge in this area. Focus on simplifying complex aspects and provide [Choose: an analogy/example/both] to make it more relatable.

Additional Instructions: [Optional: Specify if you need a brief history, applications, or implications of the topic]”

2. Practice Problems

“Could you provide a [Python code example/statistical problem solution] for a [Problem Type] task? The task should be suitable for someone with [Beginner/Intermediate/Advanced] skills. Please include comments in the code or step-by-step explanations to elucidate the thought process.

Specifics: [Optional: Include specific requirements like data sets, algorithms, statistical methods, or libraries to be used]”

3. Algorithm Explanation

“Please provide a detailed explanation of the [Algorithm Name]. This should include [Choose: its working principle, use cases, advantages, limitations, and/or comparison with other similar algorithms]. Aim the explanation towards someone with a [Basic/Intermediate/Advanced] understanding of algorithms.

Visual Aid: [Optional: Request diagrams or pseudo-code if needed]
Specific Questions: [Optional: Include any specific questions or aspects of the algorithm you want to be addressed]”

4. Code Debugging

Language/Framework: [e.g., Python, JavaScript, React]
Code Description: Briefly describe what your code is intended to do.
Issue Description: Clearly describe the problem you’re encountering (e.g., error message, unexpected output, performance issue).
Code Snippet: [Insert your code snippet here. Ensure it’s concise and relevant to the issue.]

Previous Attempts: [Optional: Mention any troubleshooting steps you’ve already taken.]
Specific Questions: [Optional: Ask specific questions related to your debugging issue.]

That’s it!

Alright, you’ve got your self-learning roadmap all set. But remember, learning these skills alone won’t instantly land you a job. Here’s what else you need to do:

  1. Firstly, dive into some real-world projects. Gather all your skills and showcase them in one spot, like GitHub or your own blog. It’s like creating a visual diary of your journey.
  2. Next, tailor your resume for each job. Make it ATS-friendly — that’s a system companies use to filter resumes. Keep applying, and trust me, your efforts will pay off.
  3. Now, let’s talk about AI. In today’s world, being AI-enabled is a must. At least get the hang of basic “prompt engineering.” It’s a skill that can set you apart.
    Learn here: The “Ultimate Guide” to Mastering the Art of Prompt Engineering: ChatGPT
  4. Communication is key, especially in data science. You’ll often need to explain complex data simply. So, sharpen your storytelling skills — they’re golden.
  5. And don’t forget about LinkedIn. Share your work actively. It’s a great way for recruiters to notice you, making your job hunt smoother.

So there you have it. Keep learning, stay curious, and you’ll definitely make it. Wishing you all the best in your journey!

If you enjoy my writings, support me:

🛒 Visit My Gumroad Shop:

My Best-selling eBook: Top 50+ ChatGPT Personas for Custom Instructions

Join my newsletter to get regular free eBooks, AI trends, and Data Science Case Studies. Subscribe now! —



Richard Warepam

Data Scientist & Writer | Google Certified Data Analyst | As a Mentor - Writes on Data Science and AI | My eBooks: