I'm in Computer Science and Data Analytics at Chapman with a 3.98 GPA.
Gold Coast Logistics — Software Engineer Intern 8 months
Built internal tools, data infrastructure, and AI agents.
Bidding engine for real-time freight auctions that generated $23K in revenue and $8.6K gross profit in two months at a 37.5% margin, versus a 15% company average. Cut bid-detection latency from 1,400ms to 450ms by swapping a polling loop for a WebSocket-triggered API fetch per bid event.
MCP server exposing 5 tools for quoting, shipment tracking, and carrier management to companywide AI agents — extending mobile quoting to 20 field reps and cutting agent quote time from ~20s to ~13s by resolving customer names and freight classes client-side instead of in the LLM.
A safe sandbox where stakeholders experiment on our real apps with live data — without touching production. I started copying the database with Docker volumes, then moved to Neon's serverless branching so each person gets a disposable full copy of live data, spun up and thrown away on demand.
Cut Google Cloud spend by $200+/month by helping migrate internal tools from Compute Engine VMs to serverless Cloud Run, and by tracing the top API expense to a location search firing a Google Maps API call on every keystroke — changing it to a single call on submit.
private repo
Real-time wearable (Polar H10 + ESP32 + Raspberry Pi 4) streaming heart-rate data to surface recovery insights unavailable in consumer apps. Re-architected from WiFi/MQTT to BLE with on-device flash logging, and 3D-printed a sensor housing.
My submission to Sigma Computing's August app-building competition, built entirely in Sigma. An AI agent matches adopters to pets: they fill out a form, then the agent chats with them about their lifestyle and combines the two to surface what matters most to them and make custom adoption recommendations. Top 10
Competed in Sigma's Builder Challenge, making an app from five years of hackathon data — projects, teams, and votes — to make organizing a hackathon easier. 2nd place — won AirPods
A Raspberry Pi scrapes SCE (no public API) and serves live pricing to a fleet of ESP32 displays spread across multiple houses — each on its own account — so no roommate questions the electric bill.
A birthday gift for my step brother, who loves planes: a self-contained ESP32-S3 e-paper display that finds the aircraft currently closest overhead and draws its route, altitude, speed, and heading, with a library of 43 unique aircraft types. All logic runs on the device with free, open APIs; no companion server.