Job
Description
Work Experience : 3+ years Salary: 21 LPA Location: Bengaluru Title : MLops Engineer Team Charter: The team in India comes with multi-disciplinary skillset, including but not limited to the following areas: Develop models and algorithms using Deep Learning and Computer Vision on the captured data to provide meaningful analysis to our customers. Some of the projects include – object detection, OCR, barcode scanning, stereovision, SLAM, 3D-reconstruction, action recognition etc. Develop integrated embedded systems for our drones – including embedded system platform development, camera and sensor integration, flight controller and motor control system development, etc. Architect and develop full stack software to interface between our solution and customer database and access – including database development, API development, UI/UX, storage, security and processing for data acquired by the drone. Integration and testing of various off the shelf sensors and other modules with drone and related software. Design algorithms related to autonomy and flight controls. Responsibilities: As a Machine Learning Ops (MLOps) engineer, you will be responsible for building and maintaining the next generation of Vimaan’s ML Platform and Infrastructure. MLOps will have a major contribution in making CV & ML offerings scalable across the company products. We are building all these data & model pipelines to scale Vimaan operations and MLOps Engineer will play a key role in enabling that. You will lead initiatives geared towards making the Computer Vision Engineers at Vimaan more productive. You will setup the infrastructure that powers the ML teams, thus simplifying the development and deployment cycles of ML models. You will help establish best practices for the ML pipeline and partner with other infrastructure ops teams to help champion them across the company. Build and maintain data pipelines - data ingestion, filtering, generating pre-populated annotations, etc. Build and maintain model pipelines - model monitoring, automated triggering of model (re)training, auto-deployment of models to producti on servers and edge devices. Own the cloud stack which comprises all ML resources. Establish standards and practices around MLOps, including governance, compliance, and data security. Collaborate on managing ML infrastructure costs. Qualifications: Deep quantitative/programming background with degree (Bachelors, Masters or Ph.D.) in a highly analytical discipline, like Statistics, Electrical,Electronics, Computer Science, Mathematics, Operations Research, etc. A minimum of 3 years of experience in managing machine learning projects end-to-end focused on MLOps. Experience with building RESTful APIs for monitoring build & production systems using automated monitoring of models and corresponding alarm tools. Experience with data versioning tools such as Data Version Control (DVC). Build and maintain data pipelines by using tools like Dagster, Airflow etc. Experience with containerizing and deploying ML models. Hands-on experience with autoML tools, experiment tracking, model management, version tracking & model training (MLflow, W&B, Neptune etc.), model hyperparameter optimization, model evaluation, and visualization (Tensorboard). Sound knowledge and experience with atleast one DL frameworks such as PyTorch, TensorFlow, Keras. Experience with container technologies (Docker, Kubernetes etc). Experience with cloud services. Working knowledge of SQL based databases. Hands on experience with Python scientific computing stack such as numpy, scipy, scikit-learn Familiarity with Linux and git. Detail oriented design, code debugging and problem-solving skills. Effective communication skills: discussing with peers and driving logic driven conclusions. Ability to perspicuously communicate complex technical/architectural problems and propose solutions for the same. How to stand out Prior experience in deploying ML & DL solutions as services Experience with multiple cloud services. Ability to collaborate effectively across functions in a fast-paced environment. Experience with technical documentation and presentation for effective dissemination of work. Engineering experience in distributed systems and data infrastructure. Show more Show less
