Job Description

• Build and train deep learning models (e.g., YOLO, RT-DETR) for object detection and classification

• Fuse data from RGB, thermal, LiDAR/ToF, IMU, and encoder sensors for real-time perception

• Implement advanced image processing (deconvolution, motion isolation, low-SNR detection)

• Work closely with hardware teams to integrate and debug sensors over GigE Vision, USB3, SPI, and I²C

• Develop embedded firmware (C/C++ or Rust) for microcontrollers and FPGAs in RTOS environments

• Create scalable data pipelines for ingestion, labeling, and training

• Optimize inference for deployment on edge platforms (GPU, FPGA)

• Build internal tools for diagnostics, performance monitoring, and auto-retraining

• Document your work and mentor other engineers on vision and embedded best practices

Qualifications

• 3–6 years of experience in computer vision, robotics perception, or sensor fusion

• Strong programming skills in C++ and Python

• Hands-on with TensorFlow, PyTorch, and real-time inference tools like TensorRT or OpenVINO

• Experience with Docker and CI/CD in ML workflows

• Background in multi-sensor calibration and synchronization

• Degree in CS, EE, Robotics, or similar (PhD a plus)

• Strong communication and cross-functional collaboration skills

• Obsessive builder mindset — you thrive solving hard technical challenges

• Experience with edge-AI optimization (quantization, pruning)

• Familiarity with embedded GPU platforms or FPGAs

• Background in safety-critical or defense systems

• Knowledge of secure coding and cybersecurity best practices