My First Steps into AI/ML: A Beginner's October 2024 Journey

Context

After two years of building a solid core programming foundation, I decided it was time to explore my initial spark: AI and Machine Learning.

What Happened

The learning curve changed:

• Started taking Andrew Ng's Machine Learning course.
• Built my first convolutional neural network (CNN) for image classification.
• Reached a ~78% accuracy rate on the model.

Unlike my chaotic Python beginnings, this transition felt more like a steady, steep climb than a frantic sprint.

Key Learnings

• AI requires a different mindset: It involves much more mathematical thinking and patience.
• Model training and debugging differ: Debugging is no longer just fixing syntax errors; it's adjusting hyperparameters and layers.
• Fundamentals over frameworks: While the urge to skip ahead to more advanced techniques was strong, I found building concrete conceptual foundations a better long-term strategy.

Takeaway

Transitioning into ML meant slowing down and trading quick dopamine hits for deep, methodical study. The vastness of the field is exciting, and patience is the only way through it.