██╗ ██████╗ ██████╗██╗ ██╗ █████╗ ███╗ ██╗
██║ ██╔═══██╗██╔════╝██║ ██║██╔══██╗████╗ ██║
██║ ██║ ██║██║ ███████║███████║██╔██╗ ██║
██║ ██║ ██║██║ ██╔══██║██╔══██║██║╚██╗██║
███████╗╚██████╔╝╚██████╗██║ ██║██║ ██║██║ ╚████║
╚══════╝ ╚═════╝ ╚═════╝╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═══╝
I build things that think. My work sits at the intersection of full-stack engineering and AI product development — where TypeScript meets Gemini, where user experience meets intelligence, and where a blank repo becomes something someone actually uses.
I'm not chasing frameworks. I'm chasing problems worth solving. Every project I ship is an attempt to answer one question: what should software actually do for people?
When I'm not writing production code, I'm exploring the edge of what AI-native interfaces can be — and pushing that edge a little further with every release.
Stop searching for answers. Start searching for the right questions.
A personalized intelligence engine — Socratic dialogue profiles you first, then Gemini 2.5 Pro synthesizes live web data into a discovery brief written specifically for you. Same query, different person, different output. Entirely by design.
TypeScriptReactGemini 2.5 ProViteLive Web Grounding
Set goals. Track progress. Actually achieve them.
A full-stack productivity platform — React frontend, Supabase backend, AI-powered goal suggestions, and real-time progress sync. Auth, database, and edge functions. Production-grade, end to end.
TypeScriptReactSupabasePostgreSQLTailwind CSSAI Suggestions
Languages C ████████████░░░░░░░░░ 60% [ process going on ]
C++ █████████████░░░░░░░░ 65% [ process going on ]
Python ████░░░░░░░░░░░░░░░░░ 20% [ process going on ]
Backend Supabase · PostgreSQL · Edge Functions
Row-Level Security · Real-time Subscriptions
AI & LLM Google Gemini 2.5 Pro · @google/genai SDK
Prompt Engineering · Context-Aware Systems
Live Web Grounding · Structured Output
Most tools are built around features. I build around outcomes.
The question I ask before writing any code is: what does the person on the other side of this interface actually need? Not what they typed into a search bar — what they're actually trying to accomplish.
That philosophy shows up in everything I ship:
- SYSTEMATIC-SEARCH doesn't just retrieve — it profiles, synthesizes, and personalizes
- GOALBOOST doesn't just track — it nudges, suggests, and adapts
- A-STARLIGHT doesn't just explain — it responds to exactly where you are in the learning journey
Good software is empathetic. It does the cognitive work so the user doesn't have to.
- Building AI-native interfaces that feel more like conversations than dashboards
- The architecture of context-aware AI systems — how to make AI genuinely useful rather than generically capable
- Supabase + real-time data patterns for collaborative, live-updating applications
- The boundary between "app" and "agent" — and what it means when that line disappears
These are projects I've built but keep out of the public spotlight — either still being refined, or intentionally not open-sourced.
Explore the cosmos. Understand the universe.
An AI-powered astronomy education platform — featuring Cosmo, a context-aware tutor that knows exactly what you're doing at any moment. Interactive sky maps, a simulated ML lab, and a narrative learning journey. Astronomy, accessible to everyone.
TypeScriptReactRechartsGemini APIViteCustom CSS Animations
I'm selectively open to opportunities where the work is ambitious, the team is sharp, and the product actually matters to the people who use it.
If that describes what you're building, let's talk.
