About the job
At Skydio, the premier drone company in the US and a global leader in autonomous flight technology, we are shaping the future of drones and aerial mobility. Our team is composed of experts in artificial intelligence, innovative hardware and software development, operational excellence, and a strong commitment to customer satisfaction. We aim to broaden the accessibility of drone technology to a diverse range of users including utility inspectors, first responders, and military personnel.
About the Position:
The essence of our autonomy system lies in its ability to learn a semantic and geometric understanding of the world through visual data. We are at the forefront of developing real-time deep learning networks to advance intelligent mobile robotics. If you are passionate about utilizing vast structured video datasets to tackle challenges in Computer Vision (CV), such as object detection, tracking, optical flow estimation, and segmentation, we want to hear from you.
As a Deep Learning Model Acceleration Engineer, you will play a pivotal role in constructing and scaling the infrastructure that underpins Skydio’s Deep Learning (DL) and AI initiatives. You will collaborate closely with our autonomy, embedded, and cloud teams to introduce new capabilities and support our deep learning efforts.
Your Impact:
Develop high-performance solutions for deep learning inference tailored to CV workloads, ensuring high throughput and low latency across various hardware platforms.
Profile CV and Vision Language Models (VLMs) to assess performance, pinpoint bottlenecks, and discover optimization opportunities while enhancing the power efficiency of deep learning inference tasks.
Design and implement comprehensive MLOps workflows for model deployment, monitoring, and retraining.
Apply your advanced Machine Learning expertise to utilize training or runtime frameworks and model efficiency tools aimed at boosting system performance.
Devise innovative methods to enhance training efficiency.
Implement GPU kernels for custom architectures and optimized inference.
Create SDKs enabling customers and external developers to construct autonomous workflows utilizing Machine Learning (ML).
Leverage your knowledge and best practices to maintain and elevate Skydio’s engineering standards.
Qualifications:
Proven hands-on experience with MLOps, ML inference acceleration/optimization, and edge deployment.
Strong foundational knowledge in deep learning frameworks and computer vision.
Ability to work collaboratively in a fast-paced environment and manage multiple projects simultaneously.
