onePost

The problem I was trying to solve

Every content creator I know has the same complaint: writing the content is the easy part. The painful part is what comes after taking one blog post and manually reformatting it for LinkedIn, Twitter, Instagram, TikTok, email, and search. That process takes 2-4 hours per piece of content, and most creators either skip platforms entirely or burn out trying to maintain presence everywhere.

I kept seeing this in Nigerian creator communities where solo entrepreneurs are doing everything themselves. Repurposing content manually was eating time they didn't have. I wanted to eliminate what I started thinking of as the "repurposing tax."

Why I chose this idea

I had a list of ideas in Notion. I kept coming back to this one. That's usually the signal.

The pain was real. The market was clear content creators, solopreneurs, marketers. The solution was technically feasible with current AI. And I felt like I could validate it quickly without needing a team or complex infrastructure.

The tech stack and why

- Next.js (App Router) — handles both frontend and API routes cleanly without a separate backend

- Groq API (Llama 3.3 70B) — fast AI generation with generous pricing

- Supabase — auth, database, storage. Fast to set up, generous free tier

- Vercel — zero-config deployment, instant preview URLs

- YouTube transcript extraction third-party library with edge-case handling layered on top

The biggest challenge I faced

Prompt engineering, honestly. It's deceptively hard to make AI write platform natively rather than just reformatting the same text. LinkedIn rewards professional storytelling and depth. TikTok hooks need to land in the first sentence. A newsletter opener is different from both.

Getting outputs that felt genuinely platform-optimized not just shorter or longer versions of the same copy took approximately 35-40 prompt iterations. I ran each version against 10 different content types, noted where it broke, refined. Repeat.

What I used AI for during the build

Everything. AI was my co-pilot for the entire build not just the product AI logic, but for debugging, writing product copy, thinking through pricing strategy, reviewing onboarding flows, and catching blind spots I'd stopped seeing after staring at the same code for a week. Building solo with an AI pair programmer is genuinely a different experience than building alone used to be. Faster, and more honest it doesn't tell you your bad ideas are good.

Current status: launched, 0 paying users

Launched today. Zero paying users. Live at https://one-post-rho.vercel.app

Free plan: 10 repurposes/month, no card required. Starter: $9/mo. Pro: $29/mo.

What I'm doing to get first users

Launching on Product Hunt today. Posting in relevant subreddits. Sharing in Nigerian tech and creator communities where I actually know people. Trying to find 10-20 people who'll use it and tell me what's broken before I try to scale anything.

No paid ads yet. I want qualitative feedback before amplifying something that might need to fundamentally change.

3 lessons from building this

1. Scope ruthlessly. I had a 20-feature roadmap. I built 8. The product is cleaner and faster for it. The other 12 features didn't need to exist to test the core value.

2. The first 10 users matter more than the next 1,000. I'm not trying to go viral. I'm trying to find the people who actually need this and have a direct conversation with them.

3. Building solo with AI tools has genuinely changed the math. Two weeks to a launched product used to require a team. That's different now. I'm not saying it's easy but the ceiling on what one person can build has shifted.

What's next on the roadmap

- Better Custom Tone Profiles improved style learning from past content

- Bulk repurpose improvements

- Analytics: tracking which platform outputs drive the most engagement

- API access for Pro users

My question for the IH community

For those who've gone from 0 → first paying users: what worked that you didn't expect? I've heard Product Hunt, cold outreach, and community posts described as the three main paths. But I'm curious what actually moved the needle for people in the first 30 days, when you had no social proof and no numbers to point to.

I'm in that exact moment right now. Any honest answer appreciated.