CS @ Rutgers. I build AI-native products that turn slow, manual workflows into grounded, LLM-powered tools that ship to real users.

A multi-tenant SaaS with an AI insights assistant — LLM-powered demand forecasting, inventory optimization, and natural-language querying over live data. Risk-scoring flags stockouts before they hit sales; a real-time dashboard and RBAC with PostgreSQL row-level security isolate every tenant.
Ingests inbound leads and auto-generates grounded, source-aware replies. n8n + LLM APIs, hardened with retry logic and rate limiting; live with a paying client within a week.
A CLI/automation platform: application-pipeline tracking, offer evaluation, AI CV generation, portal scanning, and batch processing. MIT-licensed.
I started building because I got tired of watching good work drown in slow, manual processes — so I taught myself to ship, and haven't really stopped since. Two products later, the fun part isn't the code; it's watching something you built quietly hand someone back hours of their week.
I live in the applied-AI layer — wiring LLM APIs to real data so the output is grounded and trustworthy, not just impressive in a demo. As a founder I've owned all of it: the data models, the AI, the pipelines, the UI, and the 2 a.m. reliability fixes.
What I care about: getting it right the first time, shipping fast, and building things people actually use.
Shipped two products zero → production, first paying customer live in under a week. Built multi-tenant systems with row-level security, LLM insights assistants, and real-time dashboards.
Supervised 10+ staff for programs serving 500+ participants — cut check-in time ~60% and reduced scheduling conflicts ~25% with a centralized system.
I'm actively looking for Summer / Fall 2026 internships in software and applied AI. If you're hiring — or just want to talk shop — my inbox is open.