Data Engineer Job in Bangalore at Maersk

Maersk is hiring for the role of Data Engineer!

Responsibilities of the candidate

• Ingest the world: Design and maintain ingestion frameworks for high-volume, structured and unstructured data-from operational systems, APIs, file drops, and events.
• Support streaming and batch use cases across latency windows.
• Transform at scale: Develop transformation logic using SQL, Python, Spark, and modern declarative tools like dbt or sqlmesh. You’ll handle deduplication, windowing, watermarking, late-arriving data, and more.
  Curate for trust:
• Collaborate with domain teams to annotate datasets with metadata, ownership, PII classification, and usage lineage.
• Enforce naming standards, partitioning schemes, and schema evolution policies.
• Optimize for the lakehouse: Work within a modern lakehouse architecture-leveraging Delta Lake, S3, Glue, and EMR-to ensure scalable performance and queryability across real-time and historical views.

Requirements

• Strong foundation in data engineering: You know your way around distributed systems, columnar storage formats (Parquet, Avro), data lake performance tuning, and schema evolution.
• Hands-on cloud experience: You’ve worked with AWS-native services like Glue, EMR, Athena, Lambda, and object storage (S3). Bonus if you’ve used Databricks, Snowflake, or Trino.
• Modern engineering practices: Familiarity with GitOps, containerized workflows (Docker, K8s), and CI/CD pipelines for data workflows. Experience with Terraform and IaC is highly valued.
• Programming proficiency: Fluency in Python and SQL is a must. Bonus if you’ve worked with Scala, Jinja-templated SQL, or DSL-based modeling frameworks like dbt/sqlmesh.
• Curiosity and systems thinking: You understand the tradeoffs between batch vs streaming, structured vs unstructured, cost vs latency-and you ask why before you build.
• Collaboration skills: You’ll work closely with ML engineers, platform architects, security teams, and domain data owners. Ability to communicate clearly and write clean, documented code is key.