Kevin Jimenez
Computer Engineering Graduate
Contratar a esta persona
Envía una oferta de trabajo directamente a este candidato
Sobre mí
Magna Cum Laude Computer Engineering graduate with hands-on research experience in embedded systems, computer vision, and machine learning model optimization. Proven track record across multiple appointments at Brookhaven National Laboratory designing coaxial multimodal camera systems (Event-Based & Frame-Based) and deploying compressed deep neural networks on resource-constrained edge hardware (NVIDIA Jetson, Raspberry Pi). Skilled in Python, C++, ONNX workflows, and custom sensor integration.
Experiencia
SULI Research Intern at Brookhaven National Laboratory (BNL) (2026-01 – 2026-08)
- Evaluated an advanced multimodal coaxial camera system (Prophesee EVK4 event-based camera + FLIR Blackfly S frame-based camera) utilizing a cube beam splitter and relay lens assembly to achieve near-zero baseline alignment for nuclear facility monitoring.
- Resolved multi-SDK environment incompatibilities between OpenEB (Python 3.12) and PySpin (Python 3.10) by developing an isolated interpreter acquisition launcher script (daq.sh) for synchronized data collection.
- Collected and processed ~16 minutes of synchronized, co-registered datasets across operational facility deployments (BNL Waste Management facility and Sr-90 Pilot Ground Water Treatment Plant), characterizing event noise caused by overhead lighting flicker and specular reflections off metallic surfaces.
- Leveraged the event camera's 120 dB High Dynamic Range (HDR) and microsecond temporal resolution to capture high-speed motion dynamics and underexposed regions where traditional frame-based sensors failed.
- Authored and presented two comprehensive technical research posters detailing coaxial optical alignment, software architecture, and industrial noise characterization at laboratory symposium.
Visiting Faculty Program (VFP) Summer Research Intern at Brookhaven National Laboratory (BNL) (2025-06 – 2025-08)
- Developed compressed deep neural network models for sleep stage classification targeting low-power edge deployment.
- Applied model compression techniques—including pruning, quantization, and knowledge distillation—to PyTorch models, converting them to ONNX format to execute optimized inference workflows on NVIDIA Jetson and Raspberry Pi edge platforms.
- Documented benchmarking metrics comparing inference speed, memory footprint, and classification accuracy for formal research presentations.
Undergraduate Researcher at Inter American University of Puerto Rico (2024-01 – 2025-05)
- Conducted multi-semester research on lightweight neural networks and edge computing for real-time applications.
- Benchmarked model latency and classification accuracy trade-offs across desktop environments and embedded microcontrollers.
Office Assistant at Municipality of Camuy (2019-06 – 2019-07)
- Assisted with administrative records management, data entry, and daily office operational workflows.
Educación
Bachelor of Science in Computer Engineering – Inter American University of Puerto Rico, Bayamón Campus (2025-05)
