Software Engineer (Embedded/ML Systems) at Federal Government Lab (NAVAIR) (2023-04 – 2025-09)
Spearheaded embedded machine learning team and RF mobile emitters team in model deployment, debugging, and optimization on early-stage embedded devices. Led cross-functional collaboration between government and contractor teams.
- Embedded Machine Learning Team | ML, CV, Embedded | Python, C++
- Spearheaded team in model deployment, debugging, and optimization on early-stage embedded devices, resolving key software and hardware integration issues and maintaining accuracy within 2%, despite hardware limitations using Python and C++
- Acted as primary liaison between government and contractor teams for cross-functional collaboration, streamlining communication and debugging through clear and accurate problem descriptions
- RF Mobile Emitters Team | Embedded, Full Stack, Networking | Python, C++, QML, JS
- Independently redesigned, tested and evaluated embedded hazard light algorithm in Python for 7 trailer-mounted Radio Frequency (RF) emitters, replacing a broken system with robust real-time signals for safer field operations
- Built a full-stack data pipeline in C++, QML, JavaScript (QML and JS both learned for this project) for the distributed RF emitter system on Raspberry Pi networks, cutting data review time by over 90%
- Integrated new GUI into existing GUI with intuitive playback controls to streamline analysis for the data review team
- Extended the system's 1,000+ line legacy codebase with 300+ new lines, ensuring robust TCP/UDP communication with the Command Control System (CCS) and integrating the above functionality within strict architectural constraints
- GNC Bootcamp Team | Embedded, Sensor | Python
- Automated IMU (HW123) sensor calibration via bias estimation, eliminating manual setup and speeding deployment across the team. Involved I2C and GPIO low-level hardware interfaces
