Issue 54 - My Rule of Three

This is my rule of three, which I derived from a book Practical Object-Oriented Design in Ruby for creating abstraction when it’s repeated three times. Since then I started applying this rule to everything in my day-to-day life.

In programming, I only abstract a function, class, or module if it’s being repeated three times. Because in the third attempt, you have more context and information to make the right abstraction.

For using hype products and tools, I wait at least three months for the hype to die down. You’re not going to miss anything by being three months late. I watch the news, see how it’s going but I don’t try it myself until people have used it and found it useful in this three month.

For automation, I only automate something if it happens three times. I write a script or create an agent for it. Sometimes doing it manually in a minute is faster than building a tool.

For projects, I can manage three project at a time, not more. Primary project where main income comes from, a freelance project to explore other skills, and a side project to experiment with.

For follow-ups, I do three follow-ups to get my answer. If I don’t get a response in my third attempt, I either escalate it or let it go. This applies to incoming message as well. If someone reaches out to me once than I will miss it. But if someone tries three times, I feel obligated to reply.

For health, I waited three days before seeing a doctor. For problems like gym & sports injuries, sore throat, upset stomach, muscle pain, fever, and cold, 90% of the time they get better by day three. After COVID, this health issue used caused me anxiety, but now it doesn’t.

For people, I give someone three attempts to show up on time or for a mistake. If they’re late or making same mistake for the third time, it’s a pattern. They don’t value my time and they don’t want to learn. Better to let go of those people.

For friends/colleague group, I believe you don’t need many best friends; you just need a group of three friends because there is always another perspective in the group, and you can tease one person by teaming up each time. 😂

Why Three? and NOT Four ? Because three is a really good number for finding patterns. If anything happens for...

1st time : Its Random

2nd time : Its Coincidence

3rd time : Its a Pattern.

The rule of three has become my mental framework. It helps me avoid over-engineering, overreacting, , and over-committing.

We are looking for a skilled and passionate Flutter Developer with 3–4 years of experience in mobile application development. The ideal candidate will have hands-on expertise in building high-performance, cross-platform mobile applications using Flutter and Dart

Salary Range: ₹6L – ₹7L

Want to give your users more control over their Home Screen Widgets? Imagine a weather widget where they can choose their city, or a to-do list widget displaying specific lists. This level of customization enhances user experience and makes your widgets truly personal.

It’s a new type of engineering where you orchestrate multiple tool calls, memory management, and all that stuff which we were earlier using in the available disc-based non-volatile memory of the device, but now it’s more about doing something similar but with agents.

Think digital, act analog. Thinking digital means that companies should use all the digital tools at its disposal–computers, web sites, instruments, whatever–to create great products. But companies should act analog–that is, they must remember that the purpose of innovation is not cool products and cool technologies but happy people. Happy people is a decidedly analog goal.

AI reduces the cost of production but increases the cost of coordination, review, and decision-making. And those costs fall entirely on the human.

Generative work gives you flow states. Evaluative work gives you decision fatigue.

You are collaborating with a probabilistic system, and your brain is wired for deterministic ones.

The people who waited and did nothing often ended up in a better position than the people who adopted early and had to migrate twice.

now have a hard rule: three attempts. If the AI doesn’t get me to 70% usable in three prompts, I write it myself. No exceptions. This single rule has saved me more time than any prompting technique I’ve ever learned.

It’s a shameless plug. I’ve wanted to make this video for a long time, and it was one of the most challenging to edit. Let me know if you like it or want to see more like this videos in the future.

Here is my key takeaway:

1. AI is not going to solve your organizational or human problems. Until you address the human element, AI won’t be very helpful.

2. AI is a tool. Focusing solely on a Dev’s productivity level won’t make a big impact.

3. AI adoption is extensive, but its impact within organizations is minimal.

I like the analogy of going to the moon. Just like how we found new technologies and knowledge by exploring the moon, we can do the same with AI. It won’t automate everything, but it’s a powerful tool to address organizational challenges.

Share

llamadart is a high-performance Dart and Flutter plugin for llama.cpp. It lets you run GGUF LLMs locally across native platforms and web (CPU/WebGPU bridge path).

MCP server that bridges Raycast extensions to any MCP-compatible client.

Discovers locally installed Raycast extensions, loads their tool definitions, and serves them over the Model Context Protocol via stdio or HTTP.

I am currently open for consultation part-time/full-time, specialized in mobile development with Android and Flutter. So if you are looking for someone to:

1. Build product architecture from scratch

2. Train existing developers to level up

3. Fix major bottlenecks in legacy codebase

4. Improve code quality

5. And most importantly, ship things faster

then reach out to me at info@burhanrashid52.com.