This position is a newly created role to introduce and implement machine vision to further enhance the performance and reliability of the existing robotics and automation solutions. Work and collaborate with industry experts and scientists to realize and implement end-to-end solutions.
Responsibilities
· Lead and collaborate with local research institutes to drive and derive computer vision solutions to improve overall system performance
· Working with the raw image data, research, develop, evaluate, and optimize various computer vision and deep learning models for different F&B-related automation applications
· Define and supervise the data acquisition process to build up a training database
· Exploring and visualizing data acquired to identify features, and differences in data distribution that could improve performance when deploying the model in the real world
· Analysing the errors of the model and designing strategies to overcome them
·Deploy machine vision models on edge devices after optimization to meet customer requirements and maintain them to later improve to address additional customer requirements in the future
· Defining validation strategies
· Deploying models to production
The core requirements we are looking for are as follows:
Qualifications
· MSc in Computer Science, Machine Learning, or in related fields preferred but candidates with Bachelor’s degree and relevant technical knowledge and experience in computer vision will be considered
· At least 2 years of experience in computer vision and/or deep learning for object detection and tracking along with semantic or instance segmentation either in the academic or industrial domain
· In-depth understanding and breadth of computer vision and deep learning algorithms
· Strong foundation in data structures and algorithms in Python or C++
· Proficiency with a deep learning framework and tools such as TensorFlow or Keras
· Ability to select hardware to run an ML model with the required latency including exposure to IoT technology like edge computing principles and architecture preferably for Raspberry Pi
· Experience in different model optimization techniques apart from hyperparameter tuning to reduce memory usage without hindering the performance of deploying on-edge devices
· Must be able to produce solutions independently in an organized manner and be able to work in a team
· Excellent written and verbal communication skills for effectively communicating with the team and ability to present information to a varied technical and non-technical audience.
· Willing to learn and positive, can-do attitude
About Us
https://www.rossdigital.co/
https://www.ratio.inc/
https://www.businesstimes.com.sg/garage/singapore-robotics-startup-raises-s42m-from-investors-including-fn-razer
https://www.straitstimes.com/singapore/my-psle-score-was-just-180