Applied Scientist (Computer Vision)

We are a distributed team that develops motion detection features for both Cloud and Edge environments, ensuring high performance and adaptability for diverse deployment needs. You can choose to focus on either Edge or Cloud solutions. Those interested in Edge will work on advanced computer vision algorithms and sensor-based detection methods to enable efficient real-time performance directly on devices. Those focusing on Cloud will develop scalable, high-performance motion detection solutions using cloud-based technologies, optimizing collaboration between Edge and Cloud for the best results.

You will be responsible for day-to-day research tasks, such as setting up hypotheses, gathering data for experiments, training and validating ML models, validating results, and writing a paper on your results.

• →Develop innovative machine learning algorithms. Implement advanced methods and technologies in model development
• →Analyze large datasets to identify correlations and patterns. Visualize research results for clearer interpretation
• →Tune hyperparameters and optimize algorithms. Improve models to enhance their performance and accuracy
• →Provide consultations and collaborate with colleagues from other departments
• →Report progress and project achievements to management
• →Explore and evaluate new technologies in the field of machine learning. Implement cutting-edge ideas and techniques into production processes
• →Engage in continuous self-improvement and technical skill development. Train and mentor junior colleagues to enhance the overall competence level within the team

• 4+ years in machine learning (computer vision domain)
• Practical experience in at least one of the following problems: classification, detection, segmentation
• Python3, NumPy, scikit-learn, pandas, SciPy
• Deep learning frameworks: PyTorch
• Experience in deploying machine learning models to production
• Good understanding of machine learning and deep learning concepts
• Good written and spoken English

• Practical experience with GANs, VAEs
• Probabilistic programming and bayesian framework
• Model optimization: pruning, quantization, knowledge distillation
• Basic understanding of web and client-server architecture
• asyncio, aiohttp, and other async libraries for back-end
• Basic understanding of Big Data, understanding of difference between MapReduce and in-memory processing
• Algorithms, data structures
• SQL, NoSQL
• Docker, Kubernetes, Kubeflow
• С++, Bash