Frank Feruch

Solutions Architect  ·  Principal Developer  ·  Technical Account Manager

Enterprise platform specialist with 15+ years delivering secure, scalable solutions for federal agencies, Fortune 500 companies, and higher education institutions.

I can work with any stack.

The technology is rarely the problem. I bridge the gap between business objectives and technical execution — owning client relationships, translating complex requirements into platform strategy, and then building the thing myself. I’ve delivered production systems for organizations including the U.S. Department of Justice, Johns Hopkins Press, and HP Enterprises, across Drupal, WordPress, React, Node.js, PHP, Python, and whatever else the project called for.

My experience spans enough domains that I don’t arrive with a preferred answer. Federal compliance, enterprise content architecture, performance analysis, AI-powered search, pre-sales engineering — I’ve done serious work in all of them. The stack follows the requirement.

I also use agentic AI tools like Claude Code as a core part of my workflow — not as a shortcut, but as a force multiplier. It lets me stay focused on architecture and judgment rather than boilerplate, which means better outcomes faster for anyone I work with.

I live in Walla Walla, Washington.

Walla Walla is wine country — one of the Pacific Northwest’s most celebrated appellations, with over 100 wineries within a short drive. Living here means wine is part of daily life, and naturally it became a side project.

I built a Walla Walla Wine Finder in React — an interactive map and search tool for exploring local wineries, varietals, and tasting rooms. It’s the kind of project that happens when you combine a genuine interest with a reasonable excuse to write code.

AI RAG Search — a Drupal module and a WordPress plugin.

Most search still works the same way it did in 2005: type keywords, get a list of links, read through them yourself. Retrieval-Augmented Generation changes that fundamentally.

RAG is a technique where a system first retrieves the most relevant passages from your content using semantic similarity — understanding meaning, not just matching words — then passes that context to a large language model to generate a direct, grounded answer. Instead of ten blue links, you get a real response with citations back to the source.

I built two open-source implementations to show what this looks like in practice: a contributed Drupal module published on Drupal.org, and a WordPress plugin. Both run entirely on your own server — no SaaS subscription, no data leaving your environment. The goal wasn’t to build a product. It was to show that production-quality AI search is achievable on any CMS, by any developer, without enterprise licensing fees.

Selected work.

Download my resume.

15+ years of enterprise platform experience, federal engagements, and AI integration work.

Let’s talk.

I’m available for consulting engagements. If you have an interesting problem that needs solving — regardless of the stack — I’d like to hear about it.