This is a submission for the 2026 WeCoded Challenge: Echoes of Experience
For a long time, I had nothing on paper that looked impressive enough to explain what I was trying to do.
No prestigious title. No advanced degree. No clean upward career arc that made people relax and assume I belonged.
What I did have was time, uncertainty, and a growing obsession with one question:
How do you build work that can survive skepticism when you do not yet have the life that makes people trust it?
That question changed the way I build.
At one point, the contrast in my life felt almost absurd: I was nearly two years unemployed, holding an AA on paper, while spending my days building over-documented repos, tightening assumptions, and trying to make technical claims precise enough to survive inspection.
That mismatch changed me. It made me less interested in sounding impressive and more interested in making my work defensible.
When you do not have the usual signals, you notice what people trust
Tech says it values skill. Sometimes it does. But it also values signals: a title, a degree, a known company, a polished résumé, a story that makes immediate sense.
When you do not have enough of those signals, you learn something uncomfortable: good ideas are not always enough on their own. People also need a reason to believe the idea came from someone credible.
That realization could have made me bitter. Instead, it made me more rigorous. If I could not rely on status, I had to rely on structure.
I stopped trying to sound smart and started trying to make my work inspectable
That was the shift.
I stopped asking, “How do I prove I’m capable?” and started asking better questions:
- Can someone inspect this?
- Can they follow the logic?
- Can they see how the claim connects to the evidence?
- Can the work survive contact with doubt?
That changed everything. I started treating documentation as part of the work, not decoration around it.
In one repo, that meant narrowing the scope until I could make one bounded claim honestly. In another, it meant tightening prompts, artifacts, and assumptions until the output stopped sounding clever and started sounding checkable.
I became more interested in versioned thinking than polished thinking.
What I wanted was not work that sounded impressive. I wanted work that could be checked.
One of the projects that shaped this mindset was coldplate-topobridge: not because it “proved” anything grand, but because it forced me to practice restraint. I had to narrow the claim, define the bounds, and make the artifact understandable enough that someone else could inspect it without needing to trust me first.
AI made that lesson impossible to ignore
AI did not just help me build faster. It forced me to get more serious about truth.
At first, it felt like acceleration: faster drafts, faster code, faster explanations, faster momentum. But once you work with AI long enough, you hit the wall: it can sound right long before it is right.
That changed the way I ask questions. I became less interested in what AI could generate and more interested in what it was assuming, what I had failed to specify, and what still needed verification outside the chat window.
Learning to audit AI made me better at auditing myself.
I was not just building projects. I was building a way to think under uncertainty
That is the part I wish more people talked about.
A lot of technical work is really disguised cognitive work. You are not only solving the problem in front of you. You are also managing ambiguity, self-doubt, incomplete information, and the temptation to hand-wave your way to closure.
For me, documentation stopped being a finishing touch and became part of the thinking itself. A scoped plan, a clean README, a reproducible artifact, or a well-structured prompt can all act as reality checks. They reduce the distance between what I think I am doing and what I can actually defend.
There is a strange loneliness in building before you are validated
It is one thing to build when people already believe in you. It is another to build when your effort is mostly invisible and your work has to speak before your life does.
That kind of building teaches you different lessons. It teaches you how to keep refining your standards before there is external proof that any of it will matter.
It also teaches you that some people inherit legitimacy, and some people have to assemble it manually.
What this taught me
I used to think legitimacy came first and serious work came after. Now I think a lot of people are doing serious work long before the world has decided what to call them.
Sometimes the title arrives later.
Sometimes the opportunity arrives later.
Sometimes the clean explanation arrives later.
But the rigor can start now.
That has been the biggest lesson for me: if I did not have a title to hide behind, then I needed to build work that could stand on its own.
Not perfect work. Not untouchable work. Just honest, structured, defensible work.
Work that could survive inspection.
Work that could survive doubt.
Work that could survive me.
Top comments (1)
The 'work that speaks for itself' mindset is underrated. Titles mean nothing if you can't point to something you actually built and shipped.