Yesterday, 07:55 AM
Not a "make $10k/month" post. This is the tiny, boring AI side project that actually put cash in my pocket, and why I think small beats sexy.
The project: a local restaurant near me was posting terrible Instagram Reels — phone footage, no captions, no hooks. I offered to turn each piece of raw footage into 3 short-form clips with captions, hook overlays, and a content calendar. $150/month per client. I used Whisper for captions, an LLM to write hooks and schedule suggestions, and CapCut for the actual cut (still human, takes ~10 min per batch). Four clients now. ~$600/month for maybe 3 hours a week.
Why it worked, and why a lot of the "build an AI SaaS" stuff doesn't:
- No product to build. A service, not software.
- Clients already have the pain; they don't need to be educated.
- Local = no ad spend, just walk in / cold DM.
- Retention is high because you're in their routine.
- AI is invisible to the client; they're paying for the outcome (content on their feed), not for "AI."
The lesson I keep re-learning: the easiest money with AI is often in augmenting an existing boring local business, not in building something that competes on a global app store.
What's the smallest AI side hustle that actually worked for you? Bonus points if it's boring.
The project: a local restaurant near me was posting terrible Instagram Reels — phone footage, no captions, no hooks. I offered to turn each piece of raw footage into 3 short-form clips with captions, hook overlays, and a content calendar. $150/month per client. I used Whisper for captions, an LLM to write hooks and schedule suggestions, and CapCut for the actual cut (still human, takes ~10 min per batch). Four clients now. ~$600/month for maybe 3 hours a week.
Why it worked, and why a lot of the "build an AI SaaS" stuff doesn't:
- No product to build. A service, not software.
- Clients already have the pain; they don't need to be educated.
- Local = no ad spend, just walk in / cold DM.
- Retention is high because you're in their routine.
- AI is invisible to the client; they're paying for the outcome (content on their feed), not for "AI."
The lesson I keep re-learning: the easiest money with AI is often in augmenting an existing boring local business, not in building something that competes on a global app store.
What's the smallest AI side hustle that actually worked for you? Bonus points if it's boring.


