I Taught My AI Agent to See Images
How I built a full image vision pipeline for my WhatsApp and Telegram bot — from downloading encrypted media to feeding base64 to Claude's multimodal API.
Thoughts on code, building products, and the journey of a solo founder.
ALL POSTS
How I built a full image vision pipeline for my WhatsApp and Telegram bot — from downloading encrypted media to feeding base64 to Claude's multimodal API.
Apple rejects ~40% of first-time app submissions. We decided to reject ourselves first. Found 16 issues in our React Native app that would have killed the review.
My AI agent could write blog posts but couldn't push them. The fix was a single line in .profile that was silently nuking my entire PATH.
I built a blog writing pipeline with quality checks. Two weeks later, the same problems were back. The instructions and the checklist were fighting each other.
My AI agent could search the web but couldn't see it. I gave it a real browser and it changed how we do keyword research for our blogs.
I ran my own AI slop checklist against my published blogs and found patterns I didn't even know to look for. Here's what I learned.
How I wired up Claude Code to a Telegram bot on a remote server and built a blogging pipeline that writes, reviews, and publishes posts from my phone.
The story of how I brought my AI agent, Noru, online—and the battles we fought against API gatekeepers.
Everything you need to save before wiping your machine - from obvious stuff to the files that will haunt you if you forget them.
A comprehensive guide to understanding and implementing inheritance in Java - one of the four major pillars of OOP.
A comprehensive guide to Java packages - how to create, use, and leverage advanced features for better code organization.
Understanding the Java Virtual Machine - the software that enables you to run Java programs on any device with 'Write Once, Run Anywhere' capability.