Shawn Skelly
I build AI systems that can prove they work
I build AI-powered tools and full-stack systems from architecture to deployment, and I build them to last. I spent years in quality engineering before I started building products, and that background shows in everything I ship: real test coverage, clean interfaces, and error handling that accounts for the ways things actually break. Every project started with a real problem I wanted to solve.
How I Work
Featured Projects
🌍 TALON
Land acquisition intelligence platform. Conversational AI search, 3D LiDAR visualization, ML owner classification, and deep parcel analytics across western North Carolina.
🔎 Salesforce Vector Knowledge
A Lightning Web Component that runs hybrid semantic + keyword search over Salesforce Knowledge with LLM-synthesized cited answers, at $0.0004-$0.0009 per search instead of Agentforce's $0.10/action floor.
🎯 evalharness
A retrieval eval harness built like a test framework: deterministic metrics, git-snapshotted baselines, CI gating, and an LLM that audits but never scores.
🔍 SF-Assistant
Conversational codebase assistant for Salesforce projects. Ask questions in plain English, get grounded answers with call graph context. Zero external dependencies.
🔬 RepoAudit
Codebase analysis across six dimensions: architecture, security, AI maturity, docs, portfolio, diagrams. Deterministic core plus opt-in LLM validation.
Writing
Some of this goes back to where the instincts came from, before any of it was code. Some works out how I earn the right to trust what a dataset or a model is telling me. The rest is field notes from the work itself. Different starting points, one throughline: learning to read a system well enough to know when to trust it. Some ideas run long enough to need more than one part.
I Asked Three Rival Models to Kill My Spec
before writing a line of code for a new system, i ran the spec through four adversarial review passes: my own, then three frontier models from three vendors, cold context, sequentially. 25 recorded fixes later it froze with zero open decisions. what the process caught, what it got wrong, and why the misses were still worth money.
A Recipe Is a Point, a Technique Is a Space
i learned fermentation off a dough truck at a national pizza chain, and i learned to cook without a recipe off a Wikipedia table about curry. once the fundamentals click, a recipe stops being the thing you follow and becomes one point in a space you already know how to move through.
Let's talk
Building something, hiring, or just want to compare notes on a problem? I read every message and reply to the ones worth replying to.