AI Thought Lab
A workshop for practical functional AI.
AI Thought Lab exists to document and showcase applied AI systems that solve real workflow problems. The focus is not novelty AI, prompt tricks, or abstract hype. The focus is useful systems: tools that connect AI models to real-world inputs, constraints, users, and decisions.
The site is intentionally built around proof-of-work. Projects like Swatch ID and Bolt Finder show how computer vision, OCR, retrieval, and automation can be shaped into practical tools for physical-world workflows.
Builder Background
Real-world systems thinking from R&D, CAD, and factory workflows.
Richard’s applied AI work is shaped by hands-on experience with R&D, CAD, pattern workflows, manufacturing constraints, and process improvement. That background matters because many useful AI tools fail when they are designed only as clean demos and not as practical systems.
AI Thought Lab approaches AI from the perspective of someone who cares about workflow, usability, physical environments, and whether a tool actually helps people do the work in front of them.
Working Style
Observe the real problem, build the smallest useful system, then refine.
The work starts with observation: where people lose time, where lookup breaks down, where labels or images fail, where manual process hides an opportunity for better tools. From there, the goal is to build a working prototype quickly enough to test reality.
The process is practical and iterative. Build a first version, test it against real constraints, learn what breaks, simplify where needed, and improve from there.
Professional Direction
Applied AI that bridges physical work and digital intelligence.
Richard is especially interested in AI systems that connect the physical world to searchable, automated, or decision-support workflows. That includes computer vision, visual search, OCR, vector retrieval, local AI, and human-in-the-loop tools.
AI Thought Lab is both a project archive and a professional signal: a place to show practical AI capability through real builds instead of only describing it on a resume.