Embracing the 3-Month Rule: A Pragmatic Approach to Unscalable Solutions
In the tech world, the notion of “doing things that don’t scale,” originally coined by Paul Graham, often sparks a good deal of debate. Yet, what remains largely unexplored is the practical application of this principle within software development. After dedicating eight months to building an AI podcast platform, I have developed a straightforward philosophy: any unscalable tactic I implement must endure for just three months. If it demonstrates its value, it gets refined; if not, it gets discarded.
Why the 3-Month Rule?
As engineers, we’re conditioned to focus on scalable solutions right from the outset. We gravitate towards intricate design patterns, microservices, and distributed systems—ideal for handling vast user bases. However, in a startup environment, focusing on scalability can often translate to costly delays. We tend to optimize for hypothetical users while grappling with challenges that may never arise. My three-month strategy compels me to embrace simpler, often flawed solutions that can be quickly deployed, ultimately shedding light on what my users truly need.
Current Tactical Implementations That Prove Effective
1. Consolidated Operations on a Single VM
Currently, my database, web server, background jobs, and caching are all running on a single virtual machine costing $40 a month. While this setup lacks redundancy and relies on my manual backups, the insights gained have been invaluable. Over just two months, I’ve gained a clear understanding of my actual resource consumption—my system peaks at about 4GB of RAM. The complex Kubernetes environment I nearly opted for would have left me managing virtually empty containers.
When the system encounters issues (which it has, twice), I gather concrete data on the exact points of failure. Interestingly, it’s rarely what I anticipated.
2. Hardcoded Constants for Configuration
For configuration settings, I rely on hardcoded values scattered throughout my code:
python
PRICE_TIER_1 = 9.99
PRICE_TIER_2 = 19.99
MAX_USERS = 100
AI_MODEL = "gpt-4"
This approach might seem simplistic, but it allows me to easily search my entire codebase and track any changes through Git history. I’ve only updated these variables three times in the last three months—making it far more efficient to redeploy than to build an extensive configuration service.
