I Built HustleKit AI: An AI Growth Assistant for Solo Online Sellers

Running a small online business means handling many responsibilities, from writing product descriptions to creating marketing content, understanding sales, and figuring out growth strategies. To help solo online sellers, indie makers, and creator-led businesses manage these tasks, I built HustleKit AI. This AI-powered workspace goes beyond simple text-generation chatbots by combining content creation with sales analytics and structured growth recommendations.

The platform offers several key features:

1. Product Studio for generating marketing copy and structured content using AI

2. Sales Insights for analyzing sales records, comparing performance across channels, and surfacing useful business metrics

3. AI Growth Advisor, which turns sales data into actionable recommendations through a multi-step agent workflow

4. A Business Dashboard that brings key metrics and tools into one workspace

5. Persistent Settings for storing business preferences and profile information between sessions

The core idea behind HustleKit AI is to provide independent sellers with a practical starting point for both marketing their products and deciding how to grow. The demo application can be tried at https://hustlekit-ai.onrender.com.

Technical details of the build include a full-stack TypeScript architecture, with the frontend built using React, Vite, and TypeScript, styled with Tailwind CSS and Lucide icons. The backend is powered by Node.js and Express, connected to MongoDB Atlas for persistent business data storage. Content generation and analytics are handled by Gemma 3 12B IT via OpenRouter, while Mastra orchestrates the Sales Growth Advisor workflow. Zod validates structured AI output before it's saved.

The standout feature is the Growth Advisor workflow, which separates the process into four steps: loading business data from the database, building evidence from sales metrics and business context, generating recommendations using the AI model, and validating the output against a schema before saving the report. This approach ensures that recommendations are grounded in actual business metrics.

Beyond just integrating an AI model, the project emphasized connecting model output to real application data, validating results, handling failures, and presenting everything in an intuitive user interface. Lessons learned included the importance of treating deployment and testing as integral parts of the product development process. HustleKit AI aims to make open-weight models like Gemma accessible to small businesses, providing the flexibility to understand, adapt, and extend AI-powered systems.