Role:

DarkVision’s ultrasound imaging system collects huge datasets on the order of hundreds of gigabytes that are several kilometers long where a defect may be on the order of a few mm.  It is essential that we develop algorithms for automated processing, measuring, classification and visualization of this data since manual processing is both time consuming and inconsistent. 

As a Machine Learning Engineer, you will be involved in researching, designing and implementing deep learning architectures that learn efficient generalizable representations for several tasks including: classification, regression, segmentation and object detection. You will be developing machine learning pipelines for continuous training, evaluation and deployment of deep learning models.

Our management process is light and generally milestone driven, using pseudo-agile sprints derived from our quarterly and yearly goals.  All goals are internally generated with plenty of room for feedback.

Responsibilities: (included, but not limited to)

  • Research Algorithms: Implement and train candidate deep learning architectures such as CNNs, Autoencoders and attention-based models.
  • Develop Machine Learning Pipelines: Develop cloud-based iterative pipelines for image pre-processing, data-augmentation, training, post-processing and inference.
  • Deploy and monitor ML models: Deploy ML algorithms in conjunction with other members of the Software Team for our Data Team to use and implement tools to monitor the performance.
  • Create Data collection Tools: Plan and develop tools for aiding in the collection and labelling training data while working closely with our Data team.

Expectations:

  • Smart, keen, quick, and strong desire to learn new things.
  • Bachelors or Masters degree in Computer Science, Electrical Engineering or equivalent.
  • Strong analytical skills in linear algebra, geometry, and statistics.
  • Experience developing Image Processing and conventional Computer Vision algorithms (OpenCV).
  • Experience developing Deep Learning Models for Computer Vision tasks.
  • Good working knowledge of one or more Machine Learning Frameworks (TensorFlow, Keras, PyTorch etc.)
  • Experience with Python beyond basic Machine Learning scripting (i.e., Numpy and Pandas).

Bonus Qualifications:

  • AWS services like Sagemaker, Batch, Lambda.
  • Experience training models with imbalanced datasets.
  • Experience training semantic and/or instance segmentation models.
  • Previous research experience and familiarity with latest advancements in Computer Vision.
  • Experience developing with C++.

Compensation:

  • Competitive salary or higher (Top $ for top talent)
  • Fulltime role
  • Full benefits
  • Competitive vacation allotment

Why DarkVision:

Here are 4 reason’s you’ll like working at DarkVision:

  • Well-Funded: We are well funded by solid financial backers and customers through highly differentiated technology that has already been proven out in the real world. We just raised a round from one of the largest privately held companies in the world for a large commercial expansion and new product-line development.
  • A’ Players: Our team is made up of talented, intelligent, and hardworking people. If you’re an ‘A’ player, you’ll enjoy the intellectually stimulating, challenging and respectfully competitive atmosphere.
  • Growing Quickly: We’re in the process of opening US Offices, launching several new product-lines and hiring a high-performing team of people. We are not a small start-up company and are in the process of maturing into a fast-growing medium-sized business.
  • Lots of Opportunities: We are transitioning into a much larger organization and will be expanding into a 52,000 sq ft facility in Vancouver. We are currently opening US offices, beginning in Houston.

About DarkVision:

DarkVision Technologies Inc. was founded in 2013 by a group of experienced entrepreneurs that have a track record of developing and commercializing imaging technologies for industrial markets. DarkVision has developed the world’s most advanced acoustic-based imaging platform and is packaging it into different product-lines to solve the industry’s most challenging problems.

The company has proven its first successful product-line for tubular infrastructure inspection. DarkVision’s initial product has already won multiple industry awards, been featured in top publications, and generated significant revenue. Our team consists of experts from machine vision, medical imaging, aerospace, and computer graphics sectors who are building today’s most advanced technology platform for industrial imaging and asset integrity-related inspection. We build, design, and sell the entire solution ourselves – from large sensor arrays to specialized electronics, to our visualization and analytics software.

Building on its commercial success, the company just raised a very large financing round to begin scaling the technology to new product-lines in new industries and applications. We recently began construction of a new 52,000 sq ft head office facility in Vancouver and are growing quickly. We invite you to come join our team for the exciting journey ahead as we become the global leader in industrial imaging.