She Launched an AI Product in a Weekend

Rachel Osei is a former HR consultant. She is not a developer. She has never taken a coding course, doesn't know what an API is in any technical sense, and had never built a software product before the weekend she launched Clarity — an AI tool that helps managers write performance reviews that don't sound like performance reviews.

Thirty days after that launch, she had 340 paying users and $4,980 in monthly recurring revenue.

This is the honest version of that story.

The idea Rachel had written hundreds of performance reviews over a decade in HR. She knew two things: managers hated writing them, and most of them were useless because they were written by people who hated writing them. The gap between "what I actually think about this person" and "what ends up in the review document" was where Clarity was born.

"I didn't think of it as a software product at first," she says. "I thought of it as a tool I wished I'd had for ten years. That turned out to be the right way to think about it."

The build She found Lumio through a post on X from another solo founder. She spent one evening reading the documentation, one morning connecting her OpenAI account, and one weekend building the interface. The first version had three screens: a guided prompt form, an output editor, and a save button.

"It looked like a real product. That shocked me. I kept expecting to hit the wall where I'd need a developer, and it never came."

The launch She posted in three HR communities on LinkedIn and Reddit on a Sunday evening. The post was two paragraphs — a description of the problem and a link. No Product Hunt, no press, no launch sequence. She went to sleep.

By Monday morning, 47 people had signed up. Eleven of them had completed a full review. Two had sent her direct messages asking if there was a paid version.

"I hadn't set up payments yet. That was my first mistake. I had people ready to pay and nothing to pay with."

She set up Stripe through Lumio's billing integration that Monday afternoon. By Tuesday she had her first paying customer.

The first 30 days — what actually happened

Week 1: 190 sign-ups, 23 paying users at $14/month. Biggest problem: users weren't sure how to start. She added an example review to the onboarding screen and session completion rates doubled.

Week 2: A post in a manager-focused Slack community drove 90 sign-ups in a single day. Server costs went up. Nothing broke.

Week 3: Three users emailed asking for a team plan. She hadn't considered it. She added a team tier at $49/month with five seats. Eight teams signed up in the first week.

Week 4: 340 paying users. She raised her individual plan price from $14 to $19. Churn rate: 4%.

What she'd do differently "I would have set up payments before I launched. And I would have launched a week earlier than I felt ready. The product I launched with was not the product I had planned to launch with — it was simpler, rougher, and missing two features I thought were essential. It turned out none of those features were what users cared about. I would have learned that a week sooner."

What she's building next Clarity now has a roadmap. An interview summary tool is in beta. A 360-feedback feature is planned for Q3. Rachel has one part-time contractor helping with support.

She still hasn't hired a developer.