Explorations in
AI-Powered Software
I build AI-powered applications from both sides—product thinking and systems engineering. Each project here explores a different architecture pattern, from multi-stage LLM pipelines to vision-engine hybrids.
Projects
Standalone applications, each exploring a different AI architecture pattern.
Voyagenie
voyagenie.app
AI-powered travel itineraries that actually understand what you want—from neighborhood vibes to realistic budgets, generated in seconds.
FocusChef
focuschef.app
One step at a time, with AI that actually helps when things go wrong. Type what you want to cook and get a complete, personalized recipe in seconds.
TaleForm
taleform.co
Turn cherished family memories into personalized, AI-illustrated children's books your kids will treasure forever.
ShipKit
GitHub
A product development framework that gives Claude the institutional memory and workflow scaffolding it lacks between sessions.
ChessLens
GitHub
A 2-hour prototype proving that purpose-built tools (chess engine + vision model) outperform general-purpose AI assistants at domain-specific analysis.
BidsFilter
bidsfilter.com
AI-powered procurement scanning that filters thousands of development bank tenders down to high-fit consulting opportunities, scored against your firm's profile.
Reference
Structured references for building AI-powered systems — decision-oriented, annotated, cross-referenced.
AI Capabilities
138 annotated solution patterns — what AI systems can do
System Architectures
42 blueprints — how to compose AI into complete systems
Augmentation Techniques
113 techniques — LLM augmentation and media generation
Developer Libraries
54 high-leverage library categories for AI-assisted coding