Solo-Founder Portfolio

The Solo-Founder Portfolio is my operational model for building and maintaining multiple distinct digital product brands concurrently, where AI agents handle the bulk of operational and development tasks, allowing a single operator to manage diverse revenue streams.

What it is

This approach involves developing and operating several independent product brands under one umbrella studio, Total Ventures. Instead of focusing all resources on a single venture, I distribute effort across 3-5 distinct products. Each brand targets a specific niche, has its own user base, and generates independent revenue. For example, Total Formula 1 provides F1 data and analysis, Pregnancy Power Hour offers resources for expecting parents, and Inky focuses on AI-generated content tools. The core idea is to diversify risk and revenue, ensuring that no single product's performance dictates the studio's viability. This is only feasible because AI agents manage the day-to-day operations and much of the development work.

Why I do it this way

I learned the hard way that single-product bets, while focused, carry disproportionate risk. A market shift, a platform change, or a single product's failure can halt progress entirely. By maintaining a portfolio, I spread that risk. If one product experiences a downturn, the others continue to generate revenue. This also provides more opportunities for experimentation and learning across different market segments. The ability to do this as a solo operator stems directly from my Agent-First Development strategy. My agents handle the repetitive, high-volume tasks that would typically require a team, from content generation and marketing to monitoring and basic customer support. This frees me to focus on strategic direction, system architecture, and the 5% of code that Claude Code as the Engineer doesn't get right on the first pass. It's about building durable, cash-flow positive assets without the overhead of a traditional startup.

How it works in practice

Each brand is built on a lean, serverless stack. I typically use Vercel for frontend deployment, Firebase for authentication and database, and Resend for email. Stripe handles all payment processing. My custom agent orchestration layer, VERA, integrates with Claude API and Gemini to manage workflows across these brands. For instance, an agent might monitor user feedback for Total Formula 1, generate social media updates for Pregnancy Power Hour, or draft blog posts for Inky. I use a monorepo structure for shared components and infrastructure, which simplifies maintenance. My role shifts from daily coding to architecting these agent workflows, reviewing agent-generated code, and making high-level product decisions. I prioritize Document While Shipping to ensure that the operational knowledge for each brand is captured and accessible to new agents or future human collaborators. This ensures that each product can run with minimal human intervention once launched.

Where this breaks down

While effective, this model is not without its limits. The primary constraint is my own attention. Even with robust agent systems, there are always edge cases, complex customer issues, or strategic pivots that require human judgment. If too many brands demand concurrent high-level attention, the system can strain. Onboarding a new brand still requires significant upfront architectural work and agent training. Another challenge is avoiding feature creep across products; it's easy to want to apply a successful feature from one brand to another, but this can dilute focus. Finally, while agents handle much of the execution, the initial market research and product validation for each new brand still fall squarely on me. This means the bottleneck for launching new brands is still human-driven, even if operating existing ones is largely automated.