Ml Ops

**About the Role**

We are seeking a highly skilled and experienced MLOps Engineer to join our team and play a critical role in bridging the gap between machine learning model development and production deployment. As a technology-forward company, we are dedicated to implementing cutting-edge machine learning solutions that drive our business forward. This role will play a crucial part in operationalizing artificial intelligence across our product lines, ensuring scalability, performance, and alignment with business goals.

**Responsibilities**
- Collaborate with data scientists and machine learning engineers to understand model requirements and design efficient MLOps pipelines.
- Design, develop, and implement CI/CD pipelines for machine learning models, including version control, testing, and automated deployment (utilizing tools like GitLab CI/CD, Jenkins X).
- Conduct continuous testing (CT) of models to ensure performance and identify potential issues before production deployment (using tools like Pachyderm, Metaflow).
- Deploy machine learning models (NLP, computer vision, LLM) to production on cloud platforms (AWS, GCP, Azure), ensuring scalability, security, and high availability (using tools like AWS SageMaker, Google AI Platform, Azure Machine Learning Service).
- Design and implement API endpoints for model serving using frameworks like Flask, FastAPI, and manage API gateways such as Amazon API Gateway or Kong, ensuring optimal load balancing and fault tolerance.
- Implement strategies for handling real-time and batch loads of data through APIs (using tools like Kafka, Apache Spark Streaming).
- Implement robust monitoring solutions using Prometheus, Grafana, or New Relic to track model performance, resource utilization, and operational health.
- Utilize MLOps tools like MLflow or Kubeflow for model experimentation, tracking, and deployment.
- Monitor model performance in production and implement strategies for model drift detection and mitigation (using tools like Evidently, SHAP).
- Manage multiple models in production, including scaling resources, versioning, and rollbacks (using tools like Argo Workflows, Prefect).
- Document all MLOps processes and model management activities to ensure transparency and reproducibility.
- Explore and leverage vector databases (e.g., Pinecone, Faiss, Weaviate) for efficient similarity search and retrieval tasks, particularly for NLP and recommendation systems.
- Experience with feature store platforms like Tecton or Feast for managing and serving machine learning features at scale.
- Stay up-to-date on the latest advancements in MLOps technologies and best practices.

**Qualifications**
- Master's degree in Computer Science, Engineering, or a related field (or equivalent experience).
- 5+ years of experience in MLOps or a related field.
- Proven experience in deploying Machine Learning (ML) and Deep Learning (DL) models to production.
- Experience working with structured and unstructured data.
- Proficiency in programming languages and frameworks such as Python, R, Scala, TensorFlow, PyTorch, and Scikit-learn.
- Experience in designing and implementing CI/CD pipelines for ML models with modern DevOps tools (e.g., Git, Jenkins, CircleCI) and containerization technologies (Docker, Kubernetes).
- Hands-on experience of deploying NLP, computer vision, and LLM type of models.
- Familiarity with cloud platforms like AWS, GCP, or Azure.
- Proficiency in MLOps tools like MLflow or Kubeflow.
- Strong understanding of model monitoring and continuous testing (CT) principles.
- Experience managing and scaling multiple models in production.
- Excellent communication and collaboration skills.
- Ability to work independently and as part of a team.

**Bonus Points**
- Experience with containerization technologies like Docker and Kubernetes.
- Experience with DevOps methodologies.
- Experience with infrastructure automation tools like Terraform or Ansible.
- Experience with model interpretability and explainability techniques (using tools like LIME, SHAP).
- Experience with experiment tracking tools like Neptune.ai, Weights & Biases.
- Experience with distributed training frameworks like Horovod, TensorFlow Distributed.