I built https://carbone.io (document automation) for my own use at my first company in 2012. I open-sourced it in 2017, gave a conference talk about it in 2018, and then spent very little time on marketing beyond launching a simple SaaS offering and a basic website.
It eventually became an independent company in 2021, and it's been self-funded and profitable since 2024.
I had no idea this project would one day become my full-time job!
Looking back, I regret not open-sourcing more of our internal tools.
It turned out to be one of the best ways to discover whether a product had real traction.
At the time, I struggled to justify open-sourcing our internal frameworks to my co-founders and investors.
They preferred to keep them proprietary and treat them as trade secrets.
However, most of the frameworks that remained internal have become increasingly difficult to maintain over time and now carry much greater long-term risk.
They generate maintenance costs without creating enough value in return to justify their existence.
I built https://pagey.site initially for my own self since I was creating an insane lot of small one-page sites using LLM for various utilities/calculators/explanations etc. and needed a single place to host them.
Initial version was super minimal, but then, a lot of people started using it. So, revamped it into full product and launched it. Currently hosting over 300 sites, with 10 new sites getting added daily. All free for now. I personally has 53 sites as of now.
I loved the 'Build Software Like It Needs To Last' one. I feel like it's something that could be made into a poster and hanged where I could see it first thing in the morning.
I built a question-to-sql-to dashboard tool for quick data analysis, just to avoid sharing random personal or customer xlsx documents with external AI. Released it as MIT recently:
It was able to query csv, json, parquet and xlsx files, all locally. You could also mix files of different types in single session, and, manage multiple sets of files as individual sessions that you can switch easily whenever needed.
I built https://carbone.io (document automation) for my own use at my first company in 2012. I open-sourced it in 2017, gave a conference talk about it in 2018, and then spent very little time on marketing beyond launching a simple SaaS offering and a basic website.
It eventually became an independent company in 2021, and it's been self-funded and profitable since 2024.
I had no idea this project would one day become my full-time job!
Looking back, I regret not open-sourcing more of our internal tools. It turned out to be one of the best ways to discover whether a product had real traction.
At the time, I struggled to justify open-sourcing our internal frameworks to my co-founders and investors. They preferred to keep them proprietary and treat them as trade secrets. However, most of the frameworks that remained internal have become increasingly difficult to maintain over time and now carry much greater long-term risk. They generate maintenance costs without creating enough value in return to justify their existence.
Fortunately, Carbone was an exception!
I built https://pagey.site initially for my own self since I was creating an insane lot of small one-page sites using LLM for various utilities/calculators/explanations etc. and needed a single place to host them.
Initial version was super minimal, but then, a lot of people started using it. So, revamped it into full product and launched it. Currently hosting over 300 sites, with 10 new sites getting added daily. All free for now. I personally has 53 sites as of now.
Sample stuff of mine:
1. [latency-numbers-2026] - https://latency-numbers-2026.pagey.site/
2. [NVIDIA (NVDA) — The Short / Underperformance Thesis] - https://nvidia-stock-analysis.pagey.site/
3. [Shamir's Secret Sharing] - https://shamirs-secret-sharing.pagey.site/
4. [Build Software Like It Needs To Last] - https://building-software-the-right-way.pagey.site/
5. [NPM Supply Chain Attack Techniques] - https://npm-supply-chain-attack-techniques.pagey.site/
6. [NPM Ecosystem Threat Report (May 19, 2025 - June 1, 2026)] - https://npm-supply-chain-attacks-25-26.pagey.site/
https://pagey.site
I loved the 'Build Software Like It Needs To Last' one. I feel like it's something that could be made into a poster and hanged where I could see it first thing in the morning.
I built a question-to-sql-to dashboard tool for quick data analysis, just to avoid sharing random personal or customer xlsx documents with external AI. Released it as MIT recently:
https://eatmydata.ai/
https://github.com/eatmydata-org/eatmydata
Nice... I built something same last year (not active anymore): https://www.producthunt.com/products/zenquery
It was able to query csv, json, parquet and xlsx files, all locally. You could also mix files of different types in single session, and, manage multiple sets of files as individual sessions that you can switch easily whenever needed.
I started building ToDoList to manage multiple projects on CodeProject. That was is 2003. Still working on it daily. www.abstractspoon.com