Ford Motor Company
I currently work for Ford Motor Company’s Greenfield Labs (Palo Alto, CA) as an AI & Computing Research Engineer. A general overview of my responsibilities here is as follows:
- Developing deep learning architectures and validation procedures to generate high-performance solutions for ADAS and autonomy.
- Creating computer vision and machine learning pipelines in distributed and embedded environments for processing vehicle data.
- Overseeing dataset collection, annotation, and validation for on-vehicle computer vision applications.
- Disseminating current state-of-the-art techniques and developments amongst Ford Motor Company. Additionally, developing new proof-of-concepts and intellectual property leveraging these techniques.
The primary languages and technologies I’ve utilized here are as follows:
- Python, Bash, Docker, C++
- Pytorch, Tensorflow, OpenCV, SciPy, Scikit-learn
Toyota Material Handling
I’ve worked as a Computer Vision & Robotics Engineer II in Toyota Material Handling’s Advanced R&D group (2020-2022). My main responsibilities here are listed below.
- Architecting and developing a custom perception framework to enable rapid prototyping and deployment of various computer vision and deep learning tasks.
- Enabled communication between our perception pipeline and our autonomous platforms utilizing ROS 2 Foxy to improve scene understanding and obstacle avoidance.
- Created modular interfaces with several industrial imaging devices to allow simple, scriptable configuration and easy deployment to various robot platforms.
- Created many perception proof-of-concepts using the above described perception pipeline. These include various obstacle detection and localization systems with multiple cameras.
- Implemented a multi-modality data collection routine for a gantry robot allowing rapid and autonomous data collection for various objects/obstacles.
The primary languages and technologies utilized during my time at Toyota are as follows:
- C++, CMake, Docker, Bash, Python
- OpenCV, Eigen, TensorRT, ONNX, Pytorch, Tensorflow, ROS 2
D3 Engineering
I briefly worked as a Neural Network Software Engineer at D3 Engineering in 2020. During my temporary employment I developed the initial prototype deep learning vision model for a solar panel installation robot. Contributions included implementing the model architecture, creating the training loop, embedded platform deployment and outlining future data collection efforts. A press release for recent progress on this initiative is available here.