Data Engineer — Climate & Sustainability Analytics

Ethan Caldwell

Part-time · Mid-level · UTC

This role is no longer accepting applications

Similar open roles you can apply to now:

About the role

We are an environmental science nonprofit. We monitor deforestation, biodiversity loss, and freshwater stress in tropical regions using a combination of satellite imagery, field sensor networks, and remote sensing datasets. Our scientific team produces analysis used by international conservation funders, national governments, and the IPCC working groups. We are not a tech organisation. We are a science organisation that has, over the last four years, accumulated more data than our current infrastructure can process efficiently. Our satellite data pipeline runs on scripts written by a PhD student who graduated two years ago and left comprehensive but now-dated documentation. Our field sensor database is a mess of time zones, units, and schema inconsistencies that only one person on the team fully understands. We are hiring a part-time data engineer — three days a week — to stabilise and modernise this infrastructure. We are not looking for someone to rebuild everything from scratch. We are looking for someone methodical enough to understand what we have, careful enough not to break it while improving it, and patient enough to work in an organisation where scientific rigour matters more than development velocity.

Responsibilities

Audit the existing satellite data and field sensor pipelines and produce a written assessment of stability, debt, and risk
Stabilise the highest-risk pipeline components — idempotency, error handling, and basic monitoring — without changing scientific outputs
Migrate pipeline orchestration from ad-hoc scripts to documented Airflow DAGs
Build a data quality layer for field sensor data: unit consistency, outlier detection, and gap identification
Document every data source, transformation, and schema decision in language that a new data engineer can follow without a handover call

Requirements

4+ years of data engineering with production experience on geospatial, sensor, or environmental data pipelines
Python for pipeline development and data quality automation — you write code that a scientist can read and a successor can maintain
SQL for data modelling, quality validation, and time-series extraction from sensor and monitoring databases
Apache Airflow for pipeline orchestration — you\'ve written DAGs that survive staff turnover
GCP — Cloud Storage, BigQuery, and Dataflow — or equivalent for processing satellite-derived raster and vector data at scale
Familiarity with geospatial data formats — GeoTIFF, NetCDF, GeoJSON — and basic geospatial processing is a significant advantage
Documentation as a genuine practice, not an afterthought — the single most important thing this role produces

Benefits

$43,000 – $53,000 pro-rated to three days per week
Fully remote — our team spans the UK, Kenya, Brazil, and Indonesia
Flexible hours within a 4-hour overlap with UTC
25 days annual leave pro-rated
Work on data infrastructure that directly underpins peer-reviewed scientific analysis