Computer Vision Engineer

ABOUT DURANTA & WHAT WE’RE LOOKING FOR

Duranta is an applied AI lab changing the way the world sees the world. Our flagship foundation computer vision model, Aidan, measures and understands the world in 4D, redefining how industries interact with the physical world through Artificial Spatial Intelligence. Duranta’s founding team have built unicorns, had successful exits and published groundbreaking research within the field of computer vision. We are seeking a Senior Computer Vision Engineer to join our team.

Who you'll be working with:

• Dr. Amir Zamir - maker of 4M, Taskonomy, FlexToK, winner of Best Paper award (CVPR, SIGGRAPH)

• Dr. Konrad Schindler - maker of Marigold, MOTChallenge, winner of the Frederick J. Doyle Award and Best Paper award candidates (CVPRW, ISPRS)

• Dr. Romann Bachmann - maker of MultiMAE, collaborator on MODUS, winner young investigator award

• A team of engineers and scientists with backgrounds from University of Toronto, Stanford, EPFL, ASU and NYU

You will also work closely with the founding team on deep learning research and deployment across semantic and instance segmentation, few-shot learning, 3D reasoning and generative modeling. Specifically, you will be responsible for contributing to the development of our computer vision pipelines for extracting comprehensive intelligence from optical imagery.

RESPONSIBILITIES & FOCUS AREAS

• Building the components of a state-of-the-art computer vision pipeline for our AI system

• Build and improve CV models (e.g., Mask2Former, SAM, ViT, Swin) for aerial and oblique imagery

• Run experiments at scale with deep learning frameworks like PyTorch

• Develop and maintain clean, modular ML infrastructure

• Collaborate on model deployment and inference optimizations

• Read and reproduce ideas from latest papers (CVPR, NeurIPS and other similar top-tier venues)

YOU WILL THRIVE IF YOU HAVE

• Have 5+ years experience in ML / deep learning, ideally in Computer Vision

• Have strong coding ability (Python, PyTorch)

• Have trained segmentation/detection models on real-world data at scale

• Have experience with generative modeling and large foundation models

• Have worked in startups or fast-moving teams and can wear many hats

AND FOR EXTRA CREDIT…

• Experience with high-res aerial/remote sensing imagery and LiDAR

• Experience with self-supervised and weakly-supervised learning

• Knowledge of 3D, CAD and computer graphics

• Experience with cloud computing (e.g., AWS)

DURANTA VALUES

• API = Assume Positive Intent. Our company, your relationship with co-workers and customers, and life is so much better and more fun if you follow this rule.

• Speed is the Essence. We are in a hurry to make a difference.

• Expect Peak Performance. We strive for excellence and empower those around us to do the same.

• Build something beautiful. A beautiful product, a beautiful company, and a more beautiful world.