I've been following the stories and insights here for a while, and I'm finally ready to start sharing my own journey. I'm currently working on a couple of AI-powered tools, and I thought this would be the perfect place to document the process and learn from you all.
My Path So Far: Using AI to Build AI
My development process has been a bit meta. Instead of just building AI products, I've been building my own team of AI sub-agents to help me code. I've created an "orchestrator" that manages a sequence of agents for tasks like research, architecture, coding, and testing.
For example, my refactoring workflow involves setting up pre-commit hooks, letting an LLM agent fix all the warnings, and then doing a final manual pass. It's been a fascinating experiment that has improved my code quality, even if it sometimes breaks things in unexpected ways.
The First Milestone: The Prompt Improver
After a lot of this AI-assisted coding, I'm excited to say my first product, the Prompt Improver, is now live in the Chrome Web Store. It's a simple extension designed to help users write better, more effective prompts for any LLM. Getting that "approved" email was a huge moment for me.
Of course, the journey wouldn't be complete without a hurdle. We're currently navigating a rejection from PayPal for our business account, which is the next problem to solve.
Why I'm Here
I'm really excited to be a part of this community. My goal is to share my path transparently, including the technical details of building with AI agents, the business challenges like payment gateways, and everything in between. I believe the best way to grow is to learn in public.
I'd love to connect with others who are also building in the AI space. Has anyone else had interesting experiences using AI in their own development workflow?
Thanks for reading, and I'm looking forward to the conversation.
Top comments (0)