Voxela, Inc.
Location
Delhi | India
Job description
We're building a soft real-time machine learning system that does real-time perception in the everyday world. Our system is vision based, and every facility we deploy must process terabytes of video per day from hundreds of cameras, touching on a multitude of interconnected models. We're pushing the limits of what video comprehension can achieve, and at scale. You'll be helping us solve problems that few teams have ever tackled.
As a successful DeepLearning Engineer at Voxela, you will design and build a high-quality production training pipeline and inference systems that impact our core business. All our ML Engineers contribute to the full life cycle of model development, from cross-functional data set acquisition, to training pipelines, to model design, to scaling out model serving and monitoring. If this sounds like fun, we'd love to hear from you! Responsibilities Contribute to the end-to-end development of state-of-the-art machine learning systems, from metric definition and data set creation to model deployment and monitoring Implement and Optimize state-of-the-art Detection, Pose Estimation, and Activity recognition algorithms. Solve modeling problems at the cutting edge of production machine learning Work cross-functionally with operation teams in data creation and labeling to help design our massive data sets Contribute to our low latency inference stack and develop SOTA pose estimation, action recognition, and classification. Qualifications 4+ years of experience building production software for Computer Vision systems Proficient in Python, PyTorch Proficient in at least one systems language such as Rust, C++, or Java Experience with cloud build, deployment, and orchestration tools Experience with MLOps tools such as MLFlow, Weights&Biases, Hydra or AWS Sagemaker, Experience in the implementation of Machine Learning systems and familiarity with deep learning methods Experience in computer vision, especially pose estimation, action recognition, and implementation with ultimate low latency. You communicate clearly and possess the ability to work cross-functionally across team boundaries You are able to problem solve in a flexible manner You are passionate about the entire ML lifecycle Why you might want to work with us: You're excited to work on a product that will impact almost any consumer, almost anywhere. Voxela is a remote-first company. We trust you to get your job done in the location that works best for you. We dress casually. Some of us wear slippers in the office. In current work-from-home standings, we’re super comfy. We believe in a culture of learning and want to keep building our skills, experiences, and capabilities. We offer flexible work schedules. We trust our team to know how they will do their best work. We're family-friendly. We want our teammates to focus on what they need to do when they need to. We offer very competitive compensation, including equity in Voxela, to each one of our employees.Job tags
Salary
