Data / MLOps Engineer (m/f/n)

Data / MLOps Engineer (m/f/n)
For one of our clients, we are looking for a Data / MLOps Engineer (m/f/n).
Offer
• Rate: up to 140 PLN/h net+VAT
• 100% remote
• Contract B2B (via Shimi Poland)
Project description
As a Data/MLOps Engineer within CT&C Engineering, you will be instrumental in the design, development, maintenance, and optimization of our data pipeline infrastructure and ML products. You will bridge the gap between data science and production, ensuring that our scalable data solutions provide efficient ingestion, transformation, storage, and real-time analysis.
Techinical skills
– Agile
– Apache Spark
– AWS CDK
– AWS Lambdas
– CI/CD
– ETL/ELT
– PySpark
– Python
– PyTorch
– SQL
– Training Design
Responsibilities
1\. ML & Data Infrastructure
– Deploy and maintain end-to-end ML lifecycles, focusing on automated training, deployment, and versioning.
– Build and support core MLOps components, including Feature Stores, experiment tracking, and model registries.
– Provision and manage scalable cloud infrastructure using Infrastructure as Code (IaC) (Terraform or CloudFormation).
– Develop robust CI/CD/CT (Continuous Training) pipelines to ensure seamless, repeatable production releases.
2\. Data Engineering & Pipeline Optimization
– Build high-volume ingestion and processing pipelines using Apache Spark and Python/PySpark.
– Implement data models and storage optimizations to support low-latency inference and high-performance analytics.
– Integrate automated data quality checks and observability to ensure system health and reliability.
3\. Governance & Security
– Implement proactive monitoring for model drift, data quality, and system latency.
– Maintain strict versioning for data, code, and artifacts to guarantee 100% reproducibility.
– Apply security best practices, ensuring data privacy and access control across the entire ML lifecycle.
4\. Collaborative Delivery
– Operate within an Agile framework: define technical tasks, manage tickets, and provide updates in daily stand-ups.
– Collaborate with Data Scientists and Product Owners to translate requirements into technical specifications.
– Provide clear documentation for workflows, pipelines, and architectural choices.
From the ML Ops side we really need people who
1. Can face off to data science, who are really producing notebooks only, but aren't able to implement production quality models
2. Understand different ML models and how to monitor them and the pros and cons
3. Are technically able to implement this - through pySpark and Sagemaker.
This is going to require the overall background knowledge of ML and how things work, but also decent pySpark and realistically a decent understanding of Sagemaker as well.