< Back to Blog

Building in Public at 60: What Nobody Tells You About Starting Late

TL;DR / Key Takeaways

  • Starting late means 30 years of pattern recognition most 25-year-olds simply do not have yet, and that matters more than people admit.
  • The disadvantages are real: less energy, steeper ramp on modern tooling, impostor syndrome that nobody talks about openly.
  • Financial runway from a long career is an underrated advantage. You can afford to build slowly and correctly instead of desperately.
  • Predict & Profit was built in this exact context, one working engineer's side project, and it trades real money on Kalshi today.

I turned 60 last year. I have been writing code since 1993. I built my first automated trading bot in my late 50s, sold my first copy at 59, and I am still working a full-time job while running this thing as a side project.

Nobody writes the honest version of this story. The internet is full of 27-year-olds who "built a SaaS and hit $10k MRR in 90 days." Good for them. That is not my story and it is probably not yours either. So here is the actual version.

What They Do Not Tell You About the Advantages

Everyone says experience matters. Nobody explains how.

After 30 years of building enterprise systems, I have a pattern library in my head that I could not have bought at any price when I was young. I have seen projects collapse because of technical debt nobody admitted to. I have watched three different companies nearly go under because a key system was built on tribal knowledge that lived in one person's head. I have been that person.

When I was debugging the Econ Bot and discovered that get_positions() was reading the wrong dictionary key from the Kalshi API response, the bot had been silently failing for weeks. It thought it had zero open positions. It was overtrading. That kind of silent failure is exactly the thing a junior developer does not expect and does not look for. I went straight to the API response structure because I have seen that exact class of bug before, in different systems, in different decades.

That is not skill. That is scar tissue.

There is something else. I genuinely do not care what anyone thinks about this project. Not in a defensive way. In a liberating way. I spent 30 years in corporate environments caring about optics, managing up, saying the right things in the right meetings. I do not have that gear engaged anymore. I write what I actually think. I publish the actual numbers. I admit when the bot loses. That honesty is not brave, it is just what happens when you run out of things to prove.

When you are 27 and building something, you need it to work partly because you need the validation. When you are 60, you need it to work because you are genuinely tired of working for other people and you want out. That is a different kind of motivation. It is quieter and it goes deeper.

The Financial Runway Nobody Talks About

This is the one that indie hacker Twitter mostly ignores because it feels unfair to mention.

I have a salary. I have been accumulating skills and employment for three decades. I own my tools. I am not betting rent money on whether this bot performs this month. That changes everything about how you build.

I did not cut corners to ship fast. I built the Weather Bot to use four separate ensemble sources with 164 total forecast members because that was the right architecture, not because I had to ship something that weekend. I wrote 62 unit tests for the Econ Bot's regime-change and strike-consistency systems because I had time to do it correctly. I refactored the AIGEFS module when I found the xarray truthiness bug instead of duct-taping around it.

Building without desperation makes the product better. That is the unsexy version of financial runway.

The flip side is that without desperation, you can also rationalize moving slowly indefinitely. I have to watch for that. It is easy to tell yourself you are being thorough when you are actually procrastinating.

Now the Honest Part About the Disadvantages

My energy is not what it was. That is just true. I work a full-time data engineering job at QTS Data Centers all day, come home, feed the cats, sit down at the keyboard, and I have maybe 90 minutes of sharp focus before I am done. At 30 I could code until 2am and wake up fine. Now if I do that I am useless the next day. I have had to get much more disciplined about what I actually work on, because I cannot afford to waste an evening on something that does not move the project forward.

The learning curve on modern tooling is steeper than I expected. Not because I cannot learn it. I can and I do. But there is a cost. Learning how to set up a proper GitHub Actions CI pipeline, learning the current state of the Kalshi API auth changes, integrating Open-Meteo's current response format when it changed without announcement: each one of those takes time that a developer who learned these tools recently does not spend.

I know FORTRAN. I know COBOL basics. I know technologies that are genuinely, completely irrelevant in 2026. That knowledge takes up space and it occasionally creates false confidence. "I know how systems work" is true but it does not mean I know how this specific system works today.

Then there is impostor syndrome, which I was not expecting at this age.

I thought impostor syndrome was a young person's problem. You are inexperienced, you feel like a fraud, you grow out of it. What I discovered is that it comes back in a different form when you are in a context full of people who grew up with the tools you are learning. I joined some Indie Hacker communities. Most of the active members are in their 20s and 30s. They use vocabulary I have to look up. They reference frameworks and platforms I have never heard of. They casually discuss growth hacking and SEO strategies and Discord communities as if these are obvious parts of any product launch.

I built a weather forecasting ensemble that aggregates 164 model members from four separate sources including NOAA's operational AI model and I was genuinely hesitant to post about it because I thought maybe I was doing it wrong. That is absurd. But it is real.

The way I dealt with it was to stop trying to do it their way. I do not have a Discord. I do not have a TikTok presence. I have a website, a Gumroad page, a Reddit account, and a blog where I write technical posts that I would have wanted to find when I was building this thing. 12 sales and $810 in revenue with zero paid ads, on a $97 product. That is working at a pace that makes sense for how I build.

What I Would Tell Someone in the Same Position

Build the thing you actually know something about. I spent 30 years working with data systems. The weather model ensemble I built is genuinely good because I understand data pipelines. I know how to fetch GRIB2 files with byte-range requests. I know what a race condition in a SQLite multi-thread environment looks like. I built from competence, not from a trend.

Do not try to move at a 25-year-old's pace. You are not 25. Your pace is different and your version of "fast enough" is fine. The product is better for taking the time.

Use the financial runway deliberately. It is a gift. Do not waste it on shortcuts you do not need to take.

Accept that some things will be slower and some things will be harder and that is the actual deal. The pattern recognition and the lack of ego and the discipline that comes from decades of shipping things: those do not disappear when you get tired at 10pm. They are still there in the code.

# This is the fix that took me 20 minutes to find
# because I have seen this class of bug before.
# A junior dev might have spent a week on it.

# Before: silently returning empty list on every call
positions = response.get('market_positions', [])  # wrong key

# After: correct key from actual API response structure
positions = response.get('event_positions', [])   # one line, full fix

That is the whole argument for starting late, right there. I am not faster. I am not more energetic. I do not have better tools. I just knew exactly where to look.

I am building this at 60 because I am tired of building things for other people and I want to retire to the Philippines in five years with some income that does not depend on whether some VP decides my role is redundant. That is the whole story. It is not glamorous and it is not fast. But it is mine.

That turns out to matter quite a lot.