Documenting failures is just as important as documenting success if not more. This is where I document the ideas that didn’t quite make it, the mistakes I made along the way, and the lessons I learned from failing to build products.

Each entry here is a postmortem of an idea, showing how it began, why it went wrong, and what I learned along the way.

biruni

tl;dr: Built a deterministic API for LLM datetime operations without verifying the problem first, only to find the market had already solved it with Python tool calls.

linea

tl;dr: Built a hair style try-on app for barbers and consumers, but spent too much time on plumbing before realizing the market was saturated and the business model had high churn.

calendar-agent

tl;dr: Built an AI agent to automate meeting scheduling, only to realize the problem was already solved by Google Calendar and the real issue was team habits, not tools.

fluent

tl;dr: Built a personalized language immersion app using AI-generated stories, but abandoned it after discovering a competitor already dominated the niche with a superior distribution channel.

gif-ai

tl;dr: Built an AI agent to suggest GIFs in chats, but killed it after realizing AI lacks the humor and social context needed for good GIF selection.