About the role
as an mlops engineer, you'll play a key role in operationalizing machine learning solutions that improve underwriting, claims, risk modeling, and customer experience. You will work closely with data scientists, data engineers, and actuarial teams to ensure ml models are production-ready, scalable, and resilient.
you'll be responsible for building and maintaining ml pipelines using databricks, pyspark, and spark, and automating the model lifecycle—from training to monitoring—on top of a robust cloud infrastructure.
key responsibilities
design and implement automated ml pipelines for training, testing, deployment, and monitoring of models used in insurance applications such as claims prediction, fraud detection, and policy pricing.
build scalable data workflows using pyspark and apache spark within databricks.
collaborate with data scientists and actuaries to package models and deliver reproducible, governed solutions.
implement ci/cd pipelines for ml using tools such as mlflow, azure devops, or github actions.
develop and apply techniques for data drift and model drift detection, including statistical monitoring, performance baselines, and alerts.
set up monitoring, logging, and alerting frameworks to maintain ml model reliability in production.
ensure compliance with data privacy, regulatory standards, and model governance practices required in the insurance sector.
qualifications
desired qualifications
3+ years of experience in mlops, data science engineering, or ml platform roles.
strong programming in python, with solid expertise in pyspark.
deep knowledge of spark and databricks for big data processing and scalable ml.
proficient in sql with the ability to work on complex joins and performance tuning.
experience operationalizing ml models in production (batch and real-time).
working knowledge of mlflow, docker, kubernetes, and cloud-native services (preferably azure).
proven experience in implementing and managing data drift and model drift detection using statistical and ml-based methods.
familiar with insurance data domains (e.g., claims, underwriting, loss ratio, customer churn).
understanding of data governance, model risk management, and compliance in regulated industries.
nice to have
experience with delta lake, unity catalog, or feature stores.
knowledge of data mesh, event-driven architectures, or real-time streaming.
familiarity with actuarial modeling, telematics, or fraud analytics.
certifications in azure, databricks, or mlops tools.
#j-18808-ljbffr