Grownomic AI

The Context

I left my role at Axon in September 2024 with a thesis: most small businesses know they need AI but have no idea where to start, and the tools available to them are either too generic or too complex. My co-founder and I started Grownomic to build AI-native products that solved specific, painful problems in marketing and developer tooling.

I owned the entire technical side. System architecture, frontend, backend, infrastructure, AI integrations. Everything from the first commit to production deployment.

What I Built

Agentic Google Ads Campaign Builder

The first product was an agentic campaign builder for Google Ads. The idea was simple: a small business owner gives us their website URL, and the system does the rest. It extracted business context from the site, then walked the user through offer selection, keyword research, ad group structure, campaign setup, budget allocation, and ad copy generation.

The key design decision was human-in-the-loop at every step. The AI did the heavy lifting, generated the recommendations, structured the campaign, but the user approved each decision before moving forward. This was not a "set it and forget it" tool. It was a copilot.

AI-Assisted PPC Landing Page Builder

The second product was a landing page builder designed for fast iteration under brand constraints. Clients could define design tokens (colors, typography), bring their own assets (logos, icons, testimonials), and the platform would generate conversion-optimized landing pages with managed hosting, Lighthouse performance enforcement, and integrations with analytics and ad platforms.

ChatGPT Apps & MCP Tooling

When OpenAI launched the ChatGPT Apps platform, I saw an opportunity. I built internal tooling to create MCP servers and in-chat UI components, and we secured two client contracts to build custom apps. I also prototyped an in-chat e-commerce concept that combined AI image generation with product customization for things like t-shirts, posters, and mugs.

This direction was cut short by OpenAI's December 2025 submission rules, which prohibited monetization of digital services and made distribution uncertain.

What I Learned

I shut down Grownomic in January 2026 after a year of rapid iteration and multiple pivots. The technical work was some of the most intense and rewarding I have done. But I also learned hard lessons about building on third-party platforms, the difficulty of selling to small businesses, and the importance of finding distribution before building product.

Every product we built worked. The technology was sound. What we lacked was a repeatable go-to-market motion. That lesson shaped how I think about consulting work now: I start with the business problem and the audience, not the technology.