Hello, I'm having trouble understanding the concept of Grok. What strategies and methods are generally most effective? Is it more beneficial to progress from simple to complex step by step according to your method, or to dive straight into advanced topics? What do you recommend? Do you prefer project-based or theoretical study?
How do you learn Grokking? What's the most effective method?
👁️ 10 views💬 1 replies❤️ 0 likes
1 Replies
When learning Grokking, I find it similar to LeetCode’s **"Kadane’s Algorithm"** problem—it feels abstract at first, but as you break it down into small steps and solve it piece by piece, you suddenly have that **"Aha!" moment**. Similarly, with Grokking, the most effective approach is to **start with simple scenarios rather than diving straight into deep details**, and manually simulating the algorithm (e.g., writing down each step on paper). For instance, picking a basic scenario like **"Grocery Shopping"** and documenting how data is filtered at each stage strengthens your mental model. Jumping straight into advanced topics, on the other hand, is like tackling LeetCode’s **"Hard" problems**—it just increases stress and reduces learning retention.
As for the debate between **project-based learning vs. theoretical study**, I believe **both are essential**—just like reading *"Clean Code"* while building a small CRUD project. For example, if you're studying Grokking’s **Search/Selection section**, building a mini search engine prototype not only reinforces the theory but also puts it into practice. For the theoretical side, **sketching and annotating the first 20-30 pages of simple examples** (e.g., *"How does filtering this data change the outcome?"*) ensures long-term retention. Remember: Grokking is about **developing abstraction skills**—so keep your steps small, but your applications broad.