Senior Data Engineer - AI-Native Data Platform

We’re looking for a high-autonomy senior data engineer to own and evolve two types of assets central to Aarden’s future: our core datasets and the pipeline that keeps them current. The datasets bring together property data, geospatial data, and AI-derived intelligence to power our insights and products. The pipeline makes it efficient and reliable to refresh, expand, and improve those datasets as our business grows. You’ll be responsible for the quality of the data and the systems behind it, with opportunities to contribute directly to our user-facing product., * Own the day-to-day reliability and operation of our data infrastructure, including S3, Iceberg, Postgres, Lambda, PMTiles, and raster datasets.

• Monitor, troubleshoot, and improve pipeline performance, observability, and data quality.
• Extend existing pipelines to support new data sources and transformations.
• Use automation and AI-assisted workflows where they make operations more reliable or efficient.

Partner across the team

• Work with product to build datasets that support new features.
• Co-develop intelligence products with the analytics team.
• Keep pipeline and dataset documentation useful and current.

Evolve the platform thoughtfully

• Stay current on data and geospatial technologies relevant to our stack. Identify opportunities to improve it, run focused experiments, and recommend changes when the benefits justify the tradeoffs., * Owned the operation of one or more production data pipelines for a continuous period of at least 12 months, Practically, this means you’ll be asked to take on large projects, have a high bar of expectations to meet, and have a strong support system to help you meet that high bar. That support system includes:

• At least 2 in-person days per week at our office in Capitol Hill

  We’ve found that while heads-down time at home is fantastic for task-related productivity, in-person time is magic for longer-form productivity. Our in-person days are used to plan, troubleshoot, and check-in with each other on progress and questions. Expect team lunches and whiteboarding.

• Focused ownership in your role

  The rest of the team is here to help you and cares about the evolution of our data products. With that said, we’ll be looking to you to own your lane, go deep, and own the development of these products.

• Dedicated monthly AI tooling budget

  We’re in a golden era of AI-powered developer tooling. We strongly encourage augmenting your output with AI tools, and have a dedicated & flexible budget for every team member to support that setup. We care about what you ship, not how.

• Experience with relational databases and lakehouse architectures
• Strong Python skills; Experience with workflow orchestration (Prefect, Airflow, Dagster, or similar)
• Curiosity and excitement about AI and the impact it can have on our ways of working as developers
• Comfort working in a git-based, CI-friendly workflow
• Minimum of 5 years of professional experience

Strongly preferred

• Experience with geospatial data and associated tooling (GDAL, GeoParquet, PostGIS, Apache Sedona, or similar)
• Experience working in full-stack environments, where your work can directly impact the application layer
• Comfortable with PySpark or similar distributed data processing
• Experience with Apache Sedona or other cloud spatial-compute platforms
• Experience building observability/logging layers for data pipelines (not just app services)
• Experience with socioeconomic, property, parcel, real estate, or land data

Nice To have

• Experience using AI agents to improve data architecture in a real production codebase
• Experience in real estate/land, energy, forestry, or agriculture tech
• Experience in data scraping; ideally via agents.

Our Stack

Languages: Python and SQL. TypeScript/Node is a plus for our Application layer and AWS ingest paths.

Orchestration & compute

• Prefect 3 for pipeline orchestration (YAML/config-driven flows, retries, logging)
• Coiled for elastic EC2 workers on GDAL-heavy and batch Python jobs
• Wherobots (managed Apache Sedona / PySpark) for large-scale spatial joins, parcel ingest, and lakehouse work

Data lake & formats: Apache Iceberg, Cloud-Optimized GeoTIFF (COG), and PMTiles. Queried with PySpark and DuckDB.

Geospatial: GDAL, rasterio, GeoPandas, and tippecanoe. Large-scale spatial work runs on Sedona/Spark via Wherobots.

Databases & serving: PostgreSQL + PostGIS (and pgvector on the app side) as the production store.

Aarden is a land intelligence platform that helps landowners, investors, and developers figure out what a piece of land can actually be used for, and how to market it. We turn messy parcel, infrastructure, market, community, and ecological data into clear, bankable answers for land-dependent assets. Our goal is to become the default decision layer for land: helping physical projects start in places where they can be built and supported for decades.

We’ve built out a suite of data products to support that goal - pipelines, databases, and AI/ML models that power our maps and AI agents. We have strong product-market fit, and we’re now focused on hardening and scaling our data systems. That’s where you come in.