Generative AI

1. Discovery: Before anything gets built, I want to know what's actually broken and who's living with it every day. Half of what I'm hired to build gets scoped down once we're honest about the real problem.
2. Architecture: I design for the team that has to run this after I leave. That means explainability isn't a nice-to-have bolted on at the end. It's a design constraint from day one, the same as data privacy or budget.
3. Implementation: This is where most of my own tools live, RAG pipelines, multi-agent workflows, custom automations. I build in n8n, evaluate vector databases on cost and fit rather than hype, and I write the system prompts myself.
4. Deployment and Ongoing Maintenance: AI systems drift. Models get deprecated. Costs creep. I don't hand you something and disappear. I build the monitoring in from the start so drift gets caught before it becomes a client-facing problem.

I don't just advise on AI. I build it, for my own practice and for clients.

Second Brain

At Drumline, I've led the charge on internal AI process adoption, designing and launching the Content Resonance Engine—a proprietary tool that accelerates our ability to create effective, brand-aligned blog and social media content. What once required multiple rounds of iteration now happens in a fraction of the time, freeing our team to focus on strategy and creative direction while AI handles the heavy lifting of production-level output.

I've also built an AI-powered Second Brain, which I've personally pressure-tested and now am rolling out across teams at Drumline. The Second Brain will help us operate efficiently, and effectively, allowing us to multiply the impact we have on our clients. 

Building Real AI-powered Tools

Below are just some examples working tools people can use now that show what's possible when you combine solid architecture with real-world problem-solving.