Nam Lap Vu

SeePartners – B2B SaaS affiliate marketing platform tracking traffic end-to-end (click attribution, conversion postbacks, commission payouts) with a partner cabinet and an internal admin panel; 500+ users. • Built click tracking and redirects with macro substitution, GeoIP resolution and Redis-based idempotency, designed for a 50k+ RPS ingest target. • Implemented the ClickHouse reporting layer: statistics grouped by date, offer, brand, geo and sub1–sub5, with streaming CSV/XLSX export. • Developed the pos

Embedded IoT + ML system for real-time human and vehicle detection using mmWave radar sensors with on-device machine learning; 100+ installations. • Designed and built an end-to-end embedded ML pipeline running inference directly on Raspberry Pi Pico 2W microcontrollers, classifying LD2412 24 GHz mmWave radar returns into background, human and vehicle at 5 Hz. • Engineered 58 signal-processing features from raw 14-gate radar energy data: spatial (centroid, spread, kurtosis), temporal (Hjorth parameters, sp