About the Role
As an MLOps Engineer, you will be well versed with MLOps practices and tools, working closely with data scientists/machine learning engineers to develop, operationalize and manage ML models that cater to business needs. You would have experience and expertise in various technical areas (CI/CD, Programming, Build/package, Integration, release management, monitoring, troubleshooting, etc) for delivering successful operationalization.
Responsibilities
- Develop, operationalize, and manage ML models.
- Implement model serving patterns/pipelines.
- Understand and apply Model Monitoring & Model Management principles.
- Work with large datasets and implement data pipelines, including ingestion, validation, storage, security, and processing/mining.
- Diagnose and monitor issues in MLOps using relevant tools.
Requirements
- At least 3+ years experience working as MLOps/Data Engineer (Overall, 3-8 years experience).
- At least 2+ years of Python programming experience (experience in most common libraries like numpy, pandas, mathplotlib, etc).
- Hands-on experience with scripting and coding using Python and Linux Shell.
- Must have worked as MLOps engineer in at least 2 projects.
- Must have understanding of ML Development Lifecycle (concepts).
- Must have understanding of end-to-end ML Ops Lifecycle, using relevant tools/platforms.
- Must have experience working with at least one cloud-based service (for MLOps) - AWS | Azure | Google.
- Good, hands-on experience with Linux & Containerized environments.
- Experience with Kubernetes or Docker Swarm; at least 2 projects.
- Experience with scheduling tools like Airflow, Luigi, etc. (using MLFlow/Kubeflow/ClearML).
- Understanding of automation builds (such as Jenkins/CloudBees).
- Familiarity with standard concepts and technologies used in CI/CD build, deployment pipelines, along with standard software development and release management practices.
- Experience with configuration using tools such as Chef, Ansible.
- Understanding of big data technologies (Hadoop/HDFC, Hive, Spark, Kafka, Zookeeper) and must have worked with large datasets, implemented data pipelines (2 projects).
Nice to Have
- Experience working with cloud-based services (AWS| Azure | Google).
- Aware of working in Agile teams/model and DevOps concepts, tools and practices.
- Familiarity with monitoring tools such as Prometheus, Grafana, etc. and ability to monitor and diagnose issues in MLOps.
- Good verbal and written communication skills.
- Good programming practices (coding standards, design considerations, performance tuning, etc).
- Certification in relevant topics (i.e. ML, MLOps, Cloud).
- Should be fast learner.