The 3-Step Process to Hiring a Software Engineer For Your Startup

If I ever come to need an extra hand or two for PyraMetrik, I would reject the idea of using a standard coding interview to select the best hire.

The standard coding interview (SCI) gauges coding prowess the same way an IQ test gauges intelligence.

Barely at all.

If anything, it’s a good filter for those who absolutely cannot code. But this is way too low of a bar.

And, those who ace algorithmic puzzles are not necessarily the same people who take your product to the next level with little oversight.

Solving puzzles is not a skill. Even coding is not a skill. The skill is building a world-class product. This requires not just code, but a 360-degree understanding of product and UX.

The mega-successful companies of the next decade will have little tolerance for one-dimensional skill sets.

No tech company wants to hire a different person for a product, UI/UX, code, etc. They only do it for two reasons:

1. They don’t know where else to look.
2. It’s the status quo. Within most companies, there’s this role and that. >60% bloat. Friction. Flow killer.

People separate code and product and UX — as if they were so different. They are all part of the same beast. Is a ‘proficient coder’ who doesn’t understand the product, ‘proficient’ after all?

A distributed team of coders who need management is a nightmare.

A distributed team of autonomous coders is a dream.

You want coders who don’t need managing, understand nuance, and yes, can code.

There is a way to assess one’s practical coding ability in ~ 1 hour, as it relates to building your truly unique product.

Here’s my approach. I’m not asking you to copy, as your best results come from understanding first, and apply to your unique situation second.

There are three parts to how I tackle this:

1. Coding Question — Simple But Revealing
2. Live, Specific Feature Implementation
3. Verbal, Truth-Centric Exploration

Keep in mind, this approach wouldn’t “scale” if I have to sift through 10,000 candidates, but honestly, I don’t plan to be in such a mass-hiring position anyway. That’s what I call the “growth disease” — but that’s for another time.

Part 1: Coding Question — Simple But REVEALING

First thing’s first. You hand-craft a question that can quickly determine whether the person has a basic level of understanding of data structures like arrays, maps, and sets.

If your product employs trees, throw in trees. You get the point.

Example:

https://gist.github.com/Nikhil22/d42cc6ce736daa25fc9b6cf7daac876d

Note, this question has some key qualities:

• Easy to understand. One-liners. No fluff.
• Instantly reveals your understanding of arrays, sets, and hash-maps.
• Instantly reveals your ability to write basic, clean, concise code.

If you “pass” this, it simply means you’re not incompetent. Good sign.

Step 2: Live, SPECIFIC Feature Implementation

I would carve out a section of my app, simplify it a bit, and have you implement it, either from scratch or with some starter code, depending on the time and context.

Example

https://www.loom.com/share/8d650b0200d14ac48ae69c93b004f133

This is a simple auto-complete search feature in PyraMetrik. Can you write a basic React component that successfully behaves this way? It’s a front-end skill assessment, specific to my product.

Understand — doing this forces your code to be modular. You should be able to swiftly carve out a piece of your product, throw it in a playground, and have a prospective partner work on it.

Step 3: Verbal, TRUTH-CENTRIC Exploration

Terms like Product Manager and UX Designer are fabricated, but the skills are true.

Here, I’d have a direct, honest exploration with the prospective partner to understand their outlook on building products and the domain of my product.

Our conversation reveals infinitely more about their “product” and “UX” “abilities”, than anything another so-called formal or standard measure.

There is no specific set of questions. No framework. No agenda. Whatever happens, happens.

Bottom Line

I understand that established companies that wish to run on auto-pilot prefer to mass-hire via a standardized process like SCIs. This post is not for them.

This is for the lean, uncompromising business that wishes to attract those that are absolutely necessary and effective.

Twitter

@PyraMetrik