Embedded Data Engineer for ML Pipelines & Cloud

Trainline plc; Greater London; Hybrid; GBP 70,000 - 110,000; Full time

Benefits offered by this job

Private healthcare

Dental insurance

EV scheme

Share purchase plan

Festive time off

Family-friendly benefits

Job summary

Trainline plc is hiring an Embedded Data Engineer (ML) to build scalable data pipelines and feature stores that power analytics and machine learning models. You will work with ML Engineers and Data Scientists to create reliable datasets in a cloud environment.

You will deploy cloud-native data applications on AWS, automate builds and tests via CI/CD, and ensure high performance and observability of production data pipelines. A hybrid work model is offered, with office presence required.

Qualifications

• Proficient in Python and SQL for data engineering tasks.
• Experience building data pipelines for ML workloads and model training workflows.
• Comfort with cloud data stores, data marts, and data warehouses.

Responsibilities

• Design and build scalable data pipelines, models, and feature stores for analytics and ML workloads.
• Deploy and maintain cloud-native data apps on AWS with CI/CD automation.
• Maintain data pipeline quality and reliability with observability and best practices.
• Collaborate with ML Engineers and Data Scientists to build reliable datasets for ML use cases.
• Engage with Data Engineering, Data Platform, and analytics teams to share knowledge and align on best practices.

Skills

Python

SQL

Data pipelines

Cloud AWS

Spark

Airflow

CI/CD

Terraform

Docker

Parquet/Iceberg

Tools

Docker

Terraform

GitHub Actions

Job description

Trainline plc is hiring an Embedded Data Engineer (ML) to build scalable data pipelines and feature stores that power analytics and machine learning models. You will work with ML Engineers and Data Scientists to create reliable datasets in a cloud environment.

You will deploy cloud-native data applications on AWS, automate builds and tests via CI/CD, and ensure high performance and observability of production data pipelines. A hybrid work model is offered, with office presence required.