Technical co-founder energy, without the equity.
I design and build AI solutions for startups, whether that's a new product, an integration into your existing stack, or a custom automation. I own the architecture and delivery. You focus on the business.
I'm a Fractional CTO & AI Architect who specializes in AI solutions for startups, whether that's building a new product from scratch, integrating AI into an existing system, or designing a custom automation that saves your team hours every week.
I hold a Bachelor's in Software Engineering and a Master's in Intelligent Systems. Over 8 years I've worked across 2 startups, 1 enterprise company, and consulted with 10+ clients as an AI Architect, taking products from "it works in a demo" to handling real, paying users.
Real products, real outcomes, built as a technical partner
Challenge: A solo non-technical founder with a strong vision but no technical co-founder and no product yet.
Took the idea from zero to a live product: full-stack build including AI profile generation, Stripe billing, and AWS infrastructure. Architected specifically for AI discoverability so Duuny businesses surface in ChatGPT and Claude recommendations. Shipped fast enough to support three ICP pivots without rebuilding from scratch.
Shipped to 180+ users. Featured in ChatGPT and Claude search results from launch, validating the AI-first profile strategy. Customers get ranked, location-aware results in under a second.
Challenge: An insurance startup needed to automate quote review without sacrificing the precision underwriters depend on.
Built the full PoC: from extracting structured data out of unstructured policy documents to a real-time flagging engine that catches issues before binding. The hardest part was hitting 96% extraction accuracy on messy insurance docs while keeping the workflow fast enough for a broker's day.
93.3% field extraction accuracy on a hand-labeled benchmark, with core policy sections at a clean 100%. Multi-hour manual reviews now return as background jobs in minutes with a structured summary and a prioritized, rule-referenced issue list.
Challenge: LinkedIn creators spending hours writing content that ends up sounding nothing like them.
Built from scratch as my own product: an AI writing tool that learns your tone, voice, and style to generate posts that actually sound like you. The core technical problem was the credibility scoring engine that evaluates every post before you hit publish, keeping your personal brand consistent at scale. Users are cutting content creation from 2 hours to under 10 minutes.
68 active users have generated 472 posts, averaging 3x the impressions of manually written content. Content creation time is down from 1 hour to under 5 minutes per post.
A clear, structured process so you always know what's happening and what's next.
We hop on a free 30-minute call. You walk me through your idea, your existing systems, and your timeline. I ask the hard questions so nothing gets missed.
Within 48 hours, you get a clear scope document with features, architecture plan, timeline, and a fixed price. No surprises.
I build in weekly sprints with a live demo every Friday. You see real progress every week and can steer direction as we go.
Your product goes live with documentation, deployment pipeline, and a clean codebase. I stick around for 2 weeks post-launch to squash any bugs, with the option for ongoing retainer support.
Book a free call to discuss your project. You'll get a clear scope, timeline, and quote. No surprises.