↓Skip to main content
Oliseus Blog
  1. Digitale Odyssee/

My First AI Project: A Beginner's Perspective

·451 words·3 min·
AI Journey AI-Learning Projects Python Machine-Learning Beginner Tutorial
Simon Bernbeck
Author
Simon Bernbeck
German-born AI Engineering student at International University, documenting my unconventional journey from non-CS background to AI practitioner. Currently navigating the path to becoming a digital nomad in Brazil, specifically Rio de Janeiro. Sharing learnings, travel experiences, and philosophical reflections along the way.
Table of contents

🤖 My First Steps into AI

As a non-CS graduate studying AI, this project represents my first real foray into machine learning. Here’s what I learned, the challenges I faced, and how this experience shaped my understanding of AI development.

The Project: A Simple Classification Model
#

For my first AI project, I decided to build a simple classification model using Python and scikit-learn. The goal was to predict whether a customer would purchase a product based on various features like age, income, and browsing behavior.

Why This Project?
#

Practical Application: Real-world business problem Learning Foundation: Covers fundamental ML concepts Hands-on Experience: From data preparation to model deployment

The Learning Process
#

1. Data Understanding
#

The first challenge was understanding the dataset. Coming from a non-CS background, I had to learn:

  • Data types and structures
  • Missing value handling
  • Feature engineering basics

2. Model Selection
#

I chose a Random Forest classifier because:

  • It’s robust and handles various data types well
  • Provides feature importance insights
  • Good for beginners (less hyperparameter tuning needed)

3. Implementation Challenges
#

⚠️ Key Challenge: Understanding the difference between training and test sets was initially confusing. I had to learn about overfitting and the importance of proper validation.

Results and Insights
#

The model achieved an accuracy of 85% on the test set, which was encouraging for a first attempt. More importantly, I learned:

Data Quality Matters: Clean, well-prepared data is crucial Evaluation Metrics: Accuracy isn’t everything Iterative Process: AI development is about continuous improvement

Lessons for Non-CS Students
#

What I Wish I Knew Earlier
#

  1. Start Simple: Don’t try to build complex models immediately
  2. Focus on Understanding: Code is important, but understanding the concepts is crucial
  3. Practice Regularly: Small projects build confidence and skills
  4. Join Communities: Online forums and study groups are invaluable

Resources That Helped
#

“Hands-On Machine Learning” by Aurélien Géron Coursera’s Machine Learning Course by Andrew Ng Kaggle Community for datasets and discussions

Next Steps
#

This project was just the beginning. I’m now working on:

  • More complex models (neural networks)
  • Deep learning frameworks (TensorFlow, PyTorch)
  • Real-world applications

📚 For more detailed technical discussions and code examples, check out my Substack where I dive deeper into AI topics from a student’s perspective.

Conclusion
#

My first AI project taught me that AI development is accessible to non-CS graduates. The key is persistence, continuous learning, and starting with manageable projects that build your confidence and skills.

“The best way to learn AI is to build AI. Start small, understand the fundamentals, and gradually tackle more complex challenges.” - Simon Bernbeck

This post is part of my AI learning journey. Follow along as I document my progress from non-CS graduate to AI practitioner.