About the job
At Skydio, recognized as the premier US drone manufacturer and a global pioneer in autonomous flight technology, we are dedicated to revolutionizing the future of drones and aerial mobility. Our innovative team brings together profound expertise in artificial intelligence, state-of-the-art hardware and software product development, operational excellence, and an unwavering commitment to our customers. We aim to empower a diverse range of drone users, from utility inspectors and first responders to soldiers in challenging environments and beyond.
About the Role:
The essence of our autonomy system lies in learning a semantic and geometric understanding of the world through visual data. We are at the forefront of advancing real-time deep networks to accelerate the development of intelligent mobile robots. If you are passionate about tackling complex challenges in object detection, tracking, motion prediction, flow estimation, and comprehensive scene understanding using vast amounts of structured video data, we eagerly invite you to join our team.
Your Impact:
Design and implement cutting-edge deep learning solutions that address detection, tracking, segmentation, and optical flow estimation tasks in real-time on Skydio drones.
Utilize advanced techniques in unsupervised learning, few-shot learning, and foundational models to enhance data-efficient machine learning.
Curate and enhance synthetic datasets that fuel our deep learning algorithms alongside extensive structured video data.
Refine and optimize models for low-latency performance on embedded hardware.
Characterize and evaluate the performance of our vision systems.
Research and prototype innovative methodologies.
Act as a generalist, contributing across various aspects of the software when necessary.
Ideal Candidate:
Proven hands-on experience in developing and deploying deep learning models.
Experience in curating both synthetic and real-world image datasets.
Solid foundation in software engineering with a commitment to writing clean, well-structured code (preferably in Python and C++).
Demonstrated ability to prototype, train, optimize, and deploy deep neural networks.
Strong ability to comprehend and contextualize scientific papers and literature in computer vision.
Thrives in a fast-paced, collaborative, small team environment.
