75K–95K GBP / year

Software Engineer (Data)

DataRemote — unknown
Published on 2026-11-13
These details were extracted automatically from the original listing. They may not be complete or fully up to date — it's worth checking the original job post.

About this role

Build the data and reasoning systems powering Codat’s advisory intelligence platform, initially delivered through Microsoft Copilot. The role spans data pipelines, semantic definitions, client-change signals, and LLM tools and evaluations, with responsibility for building and operating production systems.

About the company

Codat is a financial data platform that helps fintechs and banks connect to customers' accounting, commerce, and banking systems.

Stack

C#PythonSQLSparkDatabricksDelta LakeDagsterAirflow

What you'll do

  • Build and expand the semantic layer of measures, dimensions, and definitions used by the model.
  • Design systems that detect and surface meaningful changes in client data.
  • Own calculations and data structures used by the reasoning layer to produce advice.
  • Build agents, MCP servers, tests, observability, and evaluations for assistant tools.
  • Develop and operate data pipelines, services, and models from source data through to assistant responses.

What we're looking for

  • Experience building production systems from scratch and owning them in operation.
  • Experience serving data to LLMs and addressing issues such as ambiguous definitions, retrieval quality, and evaluation.
  • Backend development experience in C#, Python, or another language, with tested code.
  • Depth in at least one of SQL, Spark, Databricks/Delta Lake, Dagster/Airflow/Temporal, or dbt.
  • Experience with CI/CD, Docker, cloud infrastructure, and observability.
  • Interest in business data and setting guardrails when data cannot support a requested answer.

Nice to have

  • Experience shaping data for model queries through a metrics layer, metric views, or an MCP server.
  • Experience establishing CI/CD, Docker, cloud infrastructure, or observability practices for a team.
  • Using AI to automate repetitive parts of your own workflow.
View original job post