Lessons Learned from Building My Own SaaS in 2025/2026 (and Making $0)

At the turn of 2025 and 2026, I built a task management app for myself called Tribely. Here are my honest thoughts on building a micro-SaaS from scratch, working with AI agents, and hitting the wall of reality.

What Was My Motivation?

• Slow and bloated Jira, which we use across multiple projects at my day job.
• A desire to use something simple, lightweight, and fast.
• Wanting to learn Hotwire and Turbo in Ruby on Rails (previously, I mostly built API-only projects).
• The dream of earning some passive income ($).

Building and Writing Code in 2025

Back in 2025, I took a hybrid approach to coding. I tried to write about 50% of the code manually so I wouldn't lose my edge, while consulting the rest with AI agents. I genuinely enjoyed the process, and it helped me understand the new features introduced in Rails 8.

Problems I Didn't Anticipate

In the meantime, major competition appeared in the form of DHH and his project, Fizzy. Ironically, that made me think the task management market wasn't saturated yet and that maybe I could grab a tiny slice of the pie from the big, legacy tools. I sped up development - at this point, 90% of the code was being generated by an AI agent.

I finished the MVP (which ended up having more features than I initially planned), bought a VPS, deployed everything, and started posting about the project on my X (Twitter) account.

The result was predictable: 0 signups. Nobody (aside from a few friends) used it. After a few months, I took the app down from the server and made the code public on GitHub.

Takeaways from Building a SaaS in 2026

The SaaS market is so oversaturated right now that building something people will actually use is nearly impossible-unless you have distribution. If you don't already have established social media channels, a solid marketing strategy, or friends with huge reach, you can forget about overnight success.

The people who spent years building their personal brand and audience on social media basically won the lottery. They are in a fantastic position right now and should take full advantage of it.

What Is Worth Doing Now? (Late 2026)

• Build your personal brand (for me, this is the hardest part).
• Build your own projects using AI agents – today, AI writes 99% of my code.
• Don't overthink the tech stack – right now, the choice of technology has marginal importance compared to distribution.
• Experiment with different technologies – I come from a backend background, and currently I work with Rails, NestJS, Astro, and Rust.

Open Questions

• Does 15 years of coding experience still matter? (In my opinion, yes, especially when orchestrating AI and reviewing architecture).
• Can a solo developer still build a profitable IT product? (It has never been easier to build, yet it has never been harder to monetize).

Summary

• This is my first post here, I'm working on organizing my thoughts better, so thanks for reading!
• The good: I now understand and can comfortably use Rails' frontend stack (Hotwire/Turbo), and I have a solid project in my portfolio.
• The reality check: I made $0. I hit a wall and realized just how tough the market is right now. To make a living in IT today, you have to be more adaptable than ever pure technical skills are no longer enough.
• Three weeks ago, I open-sourced the repository on GitHub: github.com/woochaq/tribely. If anyone still looks at raw code these days, feel free to check it out - though I think we all know the answer to that question.