On Learning How to Learn

As we ride the wave of rapid technical advancements and growing business adoption of Artificial Intelligence, many of us are starting to feel the fatigue — constantly wondering what to learn, starting one topic only to switch to another a few days later, missing weeks of study (as life gets in the way), and eventually realizing after a few months that we haven’t really absorbed much at all.

Learning Artificial Intelligence and developing deep intuition around it is no easy task. Many research scientists dedicate their entire careers to this field, and it’s evolving at an incredible pace.

One thing becomes clear along the way — we need a better approach to learning. That means experimenting for ourselves, discovering what methods work best for us. It’s really about learning how to learn — so we can absorb knowledge more effectively, build real intuition, and explain complex ideas clearly to both technical and non-technical audiences.

At a physiotherapy session last week, I came across this image (below) that really caught my attention. It features three simple graphs, each representing a different approach to physical training and recovery:

1.    Top Graph: Overtraining with insufficient recovery — performance steadily declines.

2.    Middle Graph: No progressive overload — training without challenge leads to stagnation.

3.    Bottom Graph: Optimal progressive overload with proper recovery — performance steadily improves.

What struck me is how perfectly this applies to learning as well. The brain, much like a muscle, responds best to the right balance of effort and rest. The bottom graph reflects the ideal learning model: challenge yourself, then give time to recover and consolidate. That’s how we build deep understanding and sustained growth.

So how do we actually put this into practice?

“Learning how to learn” is a well-studied area, with dedicated courses and research behind it. When I began my journey into Machine Learning and Cloud Data Architecture six years ago, I went all in. But after a year of intense, scattered learning, I realised I needed a smarter, more sustainable approach.

Over time, I’ve built my own list of what to do more of — and what to do less of. I encourage you to try these out, adapt them, and create your own version that works best for you.

What to do more of?

1. Single-Tasking: Focus on one learning topic at a time. Keep a backlog list for later, but follow through with one subject until you reach a logical stopping point.
2. Daily Continuation: Continue going further and deeper on same topic for 1–2 hours daily (ideally 6 days a week). Continuation and repetition where needed builds deep understanding and intuition.
3. Deep Focus: Eliminate distractions—no phone, no social media. Dive into one topic for an hour, with a 5-minute break in between. When you're in the flow state, time disappears—and that’s when real learning happens.
4. Perseverance: You’ll get stuck—that’s part of the process. Park difficult sections and return later, but don’t abandon your progress. Be the person who finishes the book, even the boring chapters. Aim for 100% if you’re taking a course with assignments.
5. Teach & Write: Share your learning through blogs or discussions. Teaching others reveals both your strengths and gaps.
6. Hands-On Practice: Theory is valuable, but practical work cements learning. Aim for an 80:20 split—80% hands-on, 20% theory.
7. Prioritise Sleep: Aim for 7–8 hours of quality sleep.
8. Practice Meditation and Physical Activities : As we say “Sound Mind and Sound Body” and also helps improve sleep
9. Use Coffee Mindfully: Coffee can boost focus—but in moderation. Don’t overdo it.
10. Practice Niksen: Make time to do absolutely nothing. This helps shift your brain into diffused mode, allowing concepts to settle and connect—just like sleep and meditation do.

What to do less of?

1. Avoid Alcohol: It disrupts sleep, affects brain function, and gives short-term pleasure with long-term drawbacks.
2. Skip the AI Hype: Filter out the noise. Unfollow LinkedIn/Twitter accounts that sensationalise AI without substance.
3. Limit Newsletters: Most tech newsletters are marketing-heavy and distraction-prone. Be selective.
4. Don’t Chase Shiny Things: Jumping from one trend to the next erodes deep learning. Mastery comes from focus.
5. Ignore Passion Myths: Passion doesn’t have to come first. Start learning; passion often follows skill.
6. Drop Impostor Syndrome: Everyone has self-doubt—even the experts. Stay grounded and keep learning.

Readings and Acknowledgements:

• Deep Work by Cal Newport (Without Deep focus we could be just wasting time)
• Flow by Mihaly Csikszentmihalyi (Flow is the zen state / zone to be in while learning)
• Drive by Daniel H. Pink (for Drive and Curiosity)
• Grit by Angela Duckworth (for Perseverance we need when its uphill)

-Mahtab Syed, 28 May 2025, Melbourne