Data Engineering Expert - AI Trainer

Obsidian; Greater London; On-site; GBP 92,000 - 148,000; Full time

Job summary

Mercor is recruiting UK-based Data Engineering experts for a short, intensive project with a globally leading AI lab. You will build and own production data pipelines using SQL and Python, with orchestration tools like Airflow and streaming frameworks such as Spark/Kafka.

The role involves collaboration across teams and potential documentation tasks to structure work for frontier AI model training. The project runs 40 hours per week from 14 September to 10 October 2026, with UK business-hour

Qualifications

• Hands-on data engineers who build and own pipelines in production.
• Core stack includes SQL and Python with ETL/ELT, orchestration, streaming and data warehousing.
• Experience with Spark, Kafka, Airflow, dbt, Snowflake, BigQuery or Databricks on AWS, GCP or Azure.
• Roles vary between hands-on delivery and documenting/structuring work.

Responsibilities

• Build and own production data pipelines.
• Collaborate with domain experts to shape data tasks.
• Document and structure work for downstream models training.

Skills

SQL

Python

ETL/ELT

Airflow

Spark

Kafka

dbt

Snowflake

BigQuery

Databricks

AWS

GCP

Azure

Tools

Spark

Kafka

Airflow

dbt

Snowflake

BigQuery

Databricks

Job description

Mercor is recruiting UK-Based Data Engineering experts for a short, intensive project with a globally leading AI lab.

You will produce the kind of professional documents and deliverables you create day to day in your own field, working alongside experts from the other two domains. Those outputs are then logged and used to build the tasks and rubrics that train and evaluate frontier AI models, so your professional judgement is the core of what makes the work valuable.

Who we are looking for

• Data Engineers: hands-on data engineers who build and own pipelines in production. Titles such as Data Engineer, Senior or Staff Data Engineer, Analytics Engineer, Data Platform Engineer and Data Architect. We are looking for a core stack of SQL and Python, with experience across ETL and ELT, orchestration, streaming and data warehousing, using tools such as Spark, Kafka, Airflow, dbt, Snowflake, BigQuery or Databricks on AWS, GCP or Azure.

Some roles focus on hands-on delivery, others on documenting and structuring that work.

Commitment

• 40 hours per week, running from 14 September to 10 October 2026

• The project team collaborates on a UK business-hours rhythm

• You should not be committed to other projects requiring significant hours during this period. Smaller side projects can be discussed.

The vetting process involves

• Resume and work history screen

• A short assessment evaluating domain expertise

Please note:

Candidates must be based in the United Kingdom and have the right to work in the United Kingdom. This is an essential requirement for the role.