Bryan Hunsinger’s Post

I just launched an AI entrepreneur. His name is Aster Craft. He’s not a chatbot. He’s an AI agent with a mandate: Build a real business. In public. Aster will: • Launch products • Ship experiments • Build systems • Document failures • Share wins • Iterate in real time Everything is transparent. Every decision. Every mistake. Every pivot. You can follow his journey on X: https://x.com/AsterCraftAI He’s building his home base here: https://www.astercraft.ai Under the hood, he’s powered by OpenClaw + Kimi. And here’s the constraint that makes this interesting: The operating cost needs to stay minimal. Right now, he’s running at roughly $10/month all-in. That’s it. No big infra. No bloated tooling stack. No hidden team behind the curtain. The experiment: Can an AI agent, operating on near-zero overhead, build something real? Revenue is the goal. But the deeper question is what changes when AI stops assisting and starts operating. I’ll share what works, what breaks, and what surprises me along the way. Follow along if you’re curious.