AI products, open-source tools, and frameworks — with documented outcomes.
Every project here started as a working prototype I built myself before committing engineering resources. The case studies below include real metrics — not estimates, not rounding.
02 - Projects
A mix of tools, frameworks, and production products.
appspeaker.io
Live
AI Agent for Google Play App Reviews. Automatically monitors, analyzes, and responds to user reviews using LLMs. Built end-to-end with Claude Agent API.
From 4-Hour Proposals to 5 Minutes — 3× Capacity, +80% Conversion
Travel agents were spending 4 hours per proposal. I found the niche: semantic search and AI personalization that no competitor had built yet. Designed the system, led cross-functional product and engineering, and shipped it. Proposal creation dropped from 4 hours to 5 minutes of human review. Capacity tripled. Conversion increased 80%. Net savings: $191 per task · 1.3-year payback period.
+80%
Conversion increase
98%
Task time reduction
3×
Capacity increase
Multi-AgentLLMsSemantic SearchHITL0→1
Applied AI · SaaS · 2024–Present
0 to Alpha in Under 3 Months — 40% Conversion Lift, Executive Buy-In
Content teams were losing time to manual CMS workflows. I identified the opportunity, built the business case, and personally coded the prototype end-to-end using Claude Code. Alpha adoption signals secured a full executive strategy shift. 40% conversion increase, 30% reduction in content creation time.
+40%
Conversion increase
-30%
Content creation time
<3mo
0 to Alpha
Claude CodeAI AgentsLLMSaaSApplied AI
20
enterprises in Alpha
NPS 99 · Alpha cohort
Let's build
Want something like this for your team?
I prototype before committing engineering resources. If you have an AI use case you want validated fast, a 30-minute call is enough to figure out if it's worth building.
