Embracing the 3-Month Rule: A Strategic Approach to Unscalable Coding

In the tech world, the popular mantra from startup veteran Paul Graham—“Do things that don’t scale”—often makes its rounds. Yet, the discussion rarely dives into the nuances of how to effectively apply this philosophy to coding practices. After dedicating eight months to developing my AI podcast platform, I’ve paved my own path with a straightforward framework: each unscalable solution is granted a lifespan of three months. At the conclusion of this period, the hack either proves its worth and evolves into a robust solution, or it gets discarded.

The startup Mindset: Focusing on What Truly Matters

As engineers, we often feel pressured to construct scalable architectures from the outset—think microservices, distributed systems, and design patterns that cater to millions of users. However, for startups, pursuing scalability too early can amount to costly procrastination, dedicating resources to users that may not even be part of the picture yet.

The essence of my three-month rule lies in prompting me to create straightforward, albeit imperfect, code that can be rolled out quickly. This approach facilitates a direct understanding of user needs through real-world experimentation.

Key Infrastructure Strategies: Learning Through Limitations

Here are some of my current choices that may seem suboptimal but are strategically sound:

1. Centralized Operations on a Single VM

I’ve consolidated my database, web server, background jobs, and Redis onto a single $40/month virtual machine. This lack of redundancy and reliance on manual backups exposes me to the raw data of my system’s performance. After two months, I’ve gained insights into my actual resource consumption, discovering that I require just 4GB of RAM at peak times. The more elaborate container orchestration I almost implemented would have proved unnecessary and cumbersome.

2. Hardcoded Variables Across the Codebase

Configurations like pricing tiers and maximum user counts are hardcoded directly into my files. While this may seem primitive, it comes with advantages. With simple command-line utilities, I can swiftly locate any configuration value across the entire codebase, ensuring every change is logged and reviewed, albeit by myself. This approach has saved considerable engineering time—just 15 minutes of redeployment versus the week it would take to set up a dedicated configuration service.

3. SQLite: A Practical Choice for Production

Running a multi-user web application on SQLite may raise eyebrows, but my database is a mere 47MB