MLOps Professional

Job Description

Key Responsibilities: Model Hosting and Deployment: Design, develop, and maintain production-grade servers to host machine learning models. Ensure the models are deployed securely and can be accessed by APIs for multi-user interaction. Server and Infrastructure Management: Build and optimize scalable infrastructure for hosting models. Oversee the deployment and monitoring of ML models in production to ensure high availability and performance. API Development and Management: Develop and deploy RESTful APIs to allow external users and systems to interact with deployed machine learning models. Ensure APIs are performant and reliable under heavy load. Data Handling and Storage: Utilize Redis, PostgreSQL, and other databases for fast and scalable storage and retrieval of model data and predictions. Adapter model handling: Deploy various adapters for tuned models and ensure that model updates and retraining processes are seamless. Integrate new versions of models into existing systems. Documentation: Maintain comprehensive documentation for deployed models, infrastructure setup, and best practices to ensure knowledge sharing and ease of future updates. Required Skills and Experience: Model Hosting and Deployment: Strong experience with deploying and managing machine learning models in production environments. Server Infrastructure: Proficient in building and managing servers to host and serve models for scalable access. APIs: Expertise in developing and deploying RESTful APIs for model inference. Redis & Celery: Solid experience with Redis for caching and Celery for task management and background job processing. PostgreSQL: Strong knowledge of PostgreSQL for data management and querying in the context of ML applications. Model Repositories: Experience with version control and repositories for machine learning models, including integration with Git or similar platforms. CUDA & GPU: Proficient in using CUDA for accelerating machine learning workloads, especially on GPU-enabled machines. Cloud Platforms: Experience with cloud services (e.g., AWS, GCP, Azure) for deploying and managing models at scale. Preferred Skills: Familiar with CI/CD pipelines for ML models to ensure automated testing, deployment, and updates. Experience with containerization technologies such as Docker and Kubernetes for deployment and orchestration. Familiarity with frameworks like TensorFlow Serving, TorchServe, or similar for serving models in production. Knowledge of model monitoring tools and logging frameworks. Ability to work in a fast-paced, collaborative environment with cross-functional teams. Knowledge in consumer and server grade GPU to handle CUDA compatibility issues.