BioSort is developing systems that utilizes computer vision, deep learning, and biometrics to create health records for fish. Each fish is analyzed for possible disease symptoms and those which need treatment are sorted out. The result is increased fish welfare and sustainability, while lowering the cost of fish farming.
Development of BioSort’s product, iFarm, is supported heavily by the Norwegian Directorate of Fisheries and done in close partnership with one of the largest salmon farmers in the world, Cermaq, who is investing 580 million NOK in the project. This will enable collection of the largest data set for individual fish health records in the world. You will be part of a team of developers, responsible for data collection, training and verification of neural networks, as well as implementation on edge devices and cloud solutions.
BioSort currently consists of 14 employees. A multi-disciplinary development team that works to solve some of the most important challenges in fish farming, which is very technologically challenging and incredibly exciting. BioSort's success depends entirely on the knowledge and abilities of its employees and BioSort is therefore always looking for the smartest minds. The culture in the company is open, engaging and gives great freedom. BioSort is expanding the team within deep learning and software development with several new positions. Do you want to take part in this groundbreaking journey and join the BioSort family?
You will work in a team to:
- Develop and implement computer vision and machine learning techniques to identify individual fish based on visible biometrics
- Develop and implement algorithms to precisely measure the size and shape of individual fish based on Multiview geometry
- Develop and implement algorithms & pipelines to detect sea lice, wounds, and other surface anomalies on fish based on machine learning techniques
- Work closely with the embedded engineers to implement and deploy applications on high performance edge devices for real time decision making
- PhD or Master’s degree in computer science, mathematics, or equivalent combination of advanced technical education and work experience
- 3+ years of industry experience in developing products for tackling computer vision and image processing problems
- Strong foundation in computer vision, multiple view geometry, and image processing
- Strong foundation in machine learning
- Experience developing deep learning techniques for image enhancement, segmentation, saliency, object detection, object tracking, etc. in high performance environments.
- Experience with ML frameworks such as, PyTorch, Tensorflow, Nvidia Triton, etc
- Experience with programming languages such as C / C++ / Python / CUDA.
Nice to have experience:
- Expertise in developing Real Time systems
- Strong mathematical foundations in linear algebra, calculus, and probability theory
- Experience developing algorithms/applications to run on high performance embedded platforms (with GPU´s, ISPS´s, FPGA´s)
- Understanding of ML/AI Orchestration (Kubernetes, Luigi, Kubeflow, Airflow, MLFlow, etc.)
- An environment where your opinion matters, and you can make great impact
- "Startup atmosphere" with "large company stability"
- Funding secured for the next 4 years
- Nice open office, by the water, with likely the longest idea cork ever
- Electronics lab and a 20.000 liter pool for local tests with fish in sea water
- Equipment of your own choice
- Coffee from Solberg & Hansen in the office
- Several very good Cantinas, Gym, Coffee shop, and KIWI in the building
BioSort has an inclusive culture and focused on having fun at work. They love initiative and give their employees great creative freedom. BioSort’s team consist of outstanding people with extensive experience within computer vision, software, optics and mechanical product design. The founders of BioSort have long experience in product development and technology management from Tomra Systems ASA. BioSort’s ambition is to be the world leader in sensor-based optimization of fishery resources.