Data Platform Engineer

Grassroots Carbon•San Antonio, TX

About The Position

Grassroots Carbon is hiring a Data Platform Engineer to build and operate the core data infrastructure powering our soil carbon modeling work and PastureMap grazing management platform. In this role, you will integrate and serve soil measurements, satellite imagery, ranch records, and partner data through a unified database to ensure reliable, high-integrity access across the company. As the first dedicated owner of our data platform, you will drive data ingestion, dataset accessibility, and production pipeline architecture. Your immediate mission is to replace manual and ad-hoc data preparation with dependable, automated workflows that support science, operations, and rancher-facing products. You will work closely across our engineering and science teams. While Data & Soil Science handles model development and scientific validation, you will own the production pipelines and execution runtimes that put approved models into live use. You will also partner with backend engineers on application APIs and integrations, aligning key priorities with the VP of Product & Technology.

Requirements

5+ years in data engineering or data platform work, including ownership of a production data system from design through ongoing operation.
Strong Python and SQL, with maintainable code, automated tests, and practical debugging skills.
Production orchestration and deployment experience, including scheduled jobs, backfills, failure recovery, and CI/CD.
Hands-on AWS experience, infrastructure as code, and an understanding of cloud operating costs.
Experience taking ML or scientific models into production, ensuring reproducible runs, versioning, monitoring, and automated recovery.
Experience handling third-party API integrations, specifically dealing with rate limits, schema changes, and unreliable source data.

Nice To Haves

Working knowledge of geospatial data, coordinate systems, spatial joins, and tools/formats like H3, PostGIS, GDAL, GeoPandas, or cloud-optimized GeoTIFFs.
Experience implementing data access controls, secrets management, and encryption to support SOC 2 readiness or audited MRV systems.
Familiarity with model deployment tools (MLflow or SageMaker) and serverless batch processing (e.g. Modal).
Experience ingesting IoT/telemetry data from livestock devices, environmental sensors, or satellite pipelines.
BS or MS in Computer Science or similar technical studies

Responsibilities

Design and build our unified data platform from our AWS stack of S3 and PSQL with PostGIS DB, selecting the right storage and processing tools to balance workload, reliability, and cost.
Develop tested, scheduled datasets with clear lineage. Maintain consistent, versioned ranch and pasture identifiers across all internal systems.
Migrate priority PastureMap history, reconcile it to source records, and document gaps. Preserve original locations and flag records that cannot be mapped reliably.
Build and maintain reliable, automated ingestion pipelines for APIs, hardware feeds (IoT), and imagery services, replacing ad-hoc manual data extraction.
Normalize spatial and temporal incoming data from external partners and hardware, establishing clear validation and error-recovery protocols.
Build secure data services with backend engineers so PastureMap, internal tools, and approved partners can use shared datasets and model outputs.
Package, deploy, schedule, and monitor science models. Ensure code and inputs are versioned so runs are reproducible and verifiable for carbon crediting.
Provide documented datasets and accessible query tools for operations and commercial teams. Evaluate AI-assisted access where it solves a defined need and respects permissions.
Apply least-privilege access, ranch-level data isolation, secrets management, and encryption. Protect rancher and landowner information and support SOC 2 readiness.