Skip to content

// essays · 27 October 2026 · 3 min read

What you choose to attempt is the only edge left

Code, talk and ideas are cheap now. Ambition and depth are what separate people.

Code is cheap now. Talk is cheap. Ideas were always cheap.

So what’s left? Two things: what you choose to attempt, and how deep you go once you’ve started. I think those will separate people in this era. Winners and losers, if you want it blunt.

Everyone has the same tools.

The same models. The same prompts. Anyone can ask for a pipeline, a landing page or a trading bot and get something back in minutes.

So the tools can’t be the difference. You are.

“Can” beats “should.”

Dan Luu built a regex engine, and a faster fork of ripgrep, for his own workloads in a few hours. Work that would have taken him a year or more.

His point is that people keep asking whether they should. Fewer ask whether they can, now that it’s easy. It doesn’t occur to them to try.

My career is a list of things I tried.

I built my first WordPress site in 2011. By 2015 I was freelancing, hundreds of gigs, then my own agency. Then I went the other way and took jobs building payment systems at NayaPay, Funavry and Normies.co. Now I’m moving into data engineering, on top of backend, not instead of it.

Switching careers doesn’t scare me. Going from my own agency to a corporate job didn’t either. A master’s in my forties wouldn’t. Future me wouldn’t shy away from it.

It costs something, and I’m not apologizing for it.

Changing is not easy. You change your views, you leave your comfort zone, you answer people’s questions, and you have to redefine who you are.

Your life changes too. Freelancing pays differently than a corporate job, and the lifestyle that came with the first one doesn’t survive the move to the second.

Why I did it is a separate story. The point is that I chose it, and I don’t apologize for being ambitious.

Then go deeper than the first answer.

Depth is unglamorous. For me it was doing the job of the people above me. Writing documentation for others when nobody asked. Reading hundreds of pages of requirements and policy documents so compliance had no gap. Staying late so a release went out smoothly and nothing broke in production.

A model can write the code. It’s still up to a person to care whether the release is safe.

Felix Rieseberg makes the same point about AI: most people use it to do the same work faster, when they should reach higher and go deeper. Depth, for him, means using the instruments the best people in a field use, and testing against the raw data.

That matches what I’ve seen. When I rebuilt a pipeline in dbt, I planted bugs in the data on purpose. All 31 checks passed. Three rows still disappeared without a trace.

When I tested my own data loader on 143 million rows, I broke it nine ways. One row count passed with 64 damaged bytes inside a file. Depth finds the answer that was hiding, not the one you wanted.

Ambition without depth is just activity. Starting five things is easy now. I’d rather have one thing tested hard than ten that demo well.

Then show it.

Talk is cheap, so I do the work in public. I write about what I do and what I learn. I keep building open source libraries, Claude Code plugins and npm packages, and I put them on GitHub as I go.

Part of that is proof to myself that this works. Part of it is that anyone can generate polished text now. Work you can inspect, failures included, is rarer, and that’s what makes it believable.

The tools get cheaper for everyone. Ambition doesn’t.

p.s. If this sounds like a pep talk, check the links above. The proof is in the work.

Muhammad Mairaj is a Senior Backend Engineer with 8 years building payment and data systems. He writes about data engineering, measured. More about him.