Cover is an AI security company developing concealed weapon detection systems. Cover's imaging technology scans students for concealed weapons in K-12 schools in the United States. Our goal is to deter school shootings by identifying concealed weapons inside of bags and underneath clothing. We are headquartered in Sunnyvale, CA with offices in Pasadena, CA and require 5 days/week in-office work.
In order to scale to the 130,000 K-12 schools in the United States, we will need to detect weapons with fully autonomous AI models. We are looking for a deep learning engineer to take in sparse point clouds and output weapon models with low false positives. Your goal is to design weapon detection models with low latency and high accuracy to prevent school shootings.
Responsibilities:
- Research, design, implement, optimize and deploy deep learning models that advance the state of the art in autonomous weapons detection models
- Operate with a commercial mindset to ship working product that can detect weapons at K-12 schools in the U.S.
- Train machine learning and deep learning models on a computing cluster to carry out visual recognition tasks, including weapon segmentation and detection
- Review deep learning code and research papers, implement models and algorithms, tailor them to our specific use cases for school weapon detection, enhance internal metrics, and collaborate with downstream engineers to efficiently integrate neural networks into our scanning system
- Enhance deep neural networks and their related preprocessing and postprocessing code to ensure efficient execution on an embedded device
Requirements:
- Experience with PyTorch, or at least another major deep learning framework such as TensorFlow, MXNet
- Deep comprehension of the foundational principles of deep learning, including layer architecture, backpropagation, and other essential concepts
- Strong desire to help ship AI systems that can prevent school shootings