The 3-Month Rule: An Ingenious Framework for Unconventional Coding Strategies

In the world of technology, Paul Graham’s mantra, “Do things that don’t scale,” resonates strongly, yet the practical application of this wisdom in programming is often overlooked. As I recently embarked on the ambitious journey of developing my AI podcast platform over the past eight months, I established a straightforward principle: any temporary, non-scalable solution is limited to a three-month lifespan. After this period, the solution either demonstrates its worth and is refined into a robust feature, or it will be retired.

As engineers, we’re conditioned to prioritize scalable solutions from the outset—embracing design patterns, microservices, and distributed systems that support thousands, if not millions, of users. However, this “big company” mentality can hinder progress in a startup environment where scalable code may simply result in costly delays. By adhering to my three-month rule, I compel myself to write straightforward, sometimes imperfect code that is prompt to deploy and yields insights into what users truly require.

Current Infrastructure Hacks: Smart Choices for Learning

1. Single VM Setup

Currently, my entire infrastructure operates on a single virtual machine (VM) that costs just $40 a month, hosting everything from the database to background jobs without any redundancy. While this might seem like a reckless approach, it has granted me invaluable insights concerning my actual resource consumption in just two months. My so-called “AI-heavy” platform only requires 4GB of RAM at peak. Had I gone ahead with an elaborate Kubernetes setup, I would have ended up managing empty containers. And when issues arise (which they have), I gain firsthand knowledge about the true causes of failure—insights that are often surprising.

2. Hardcoded Constants

In my code, configuration values are hardcoded like this:

plaintext
PRICE_TIER_1 = 9.99
PRICE_TIER_2 = 19.99
MAX_USERS = 100
AI_MODEL = "gpt-4"

This means no external configuration files or environment variables. Any modification requires a redeployment, which might appear limiting, but there’s a silver lining. By running a simple search across my codebase, I can track changes in configuration quickly, all within seconds. In the three months that I have been operating this way, I have made only three changes in configurations, resulting in a mere 15 minutes of redeployment versus the 40 hours