MLOps Engineer

About Grassroots Carbon

Grassroots Carbon is the leading grasslands restoration and soil carbon storage company that partners with landowners to implement and scale regenerative land management practices. In addition to enhancing soil health, promoting biodiversity, and improving water quality, these regenerative practices have tremendous potential to combat climate change by drawing down large quantities of atmospheric CO2 into the soil. Grassroots Carbon is proud to have partnered with ranchers across 1.6 million acres in 22 states to implement practices that restore grasslands, improve bird habitats, build soil health, and drive nature-based soil organic carbon drawdown through the healthy soils.

Built on a foundation of scientific rigor, quality, and transparency, Grassroots Carbon has built strong partnerships with Audubon Conservation Ranching, Texas Agricultural Land Trust, Understand Ag, and Colorado State University's Soil Carbon Solutions Center while generating high-quality soil carbon drawdown credits for leading corporations, including Nestle, Microsoft, Shopify, Marathon Oil, H-E-B, Olipop, and Urban Villages, to offset their carbon impact and reach their sustainability goals.

Grassroots Carbon is proud to be a portfolio company of Soilworks Natural Capital

Grassroots Carbon is seeking a hands-on AI/ML Ops Engineer to build and maintain the data infrastructure powering our carbon measurement, rancher tools, and emerging AI-native applications.

You’ll work at the intersection of our Data Science and Product teams, translating experimental models into production systems while building the connectors that enable real-time data flows across our platform. This role is critical to our 2026 goals: delivering AI-native ranch management tools, scaling our MRV (Measurement, Reporting, Verification) capabilities, and enabling third-party hardware integrations that put actionable intelligence directly into ranchers’ hands.

We’re looking for someone who thrives on building reliable infrastructure, has strong opinions about data quality, and can work autonomously in a fast-moving environment where the roadmap is ambitious and the problems are genuinely novel.

Data Pipeline Infrastructure

• Design, build, and maintain scalable ETL pipelines for soil carbon data, satellite imagery, weather data, and ranch management inputs
• Productionize data science team outputs, transforming research models (process-based and machine learning) into reliable, monitored production systems
• Build and maintain data validation, monitoring, and alerting systems that catch issues before they impact downstream users
• Own data quality and lineage across the platform, ensuring our carbon calculations maintain the scientific rigor our protocols require

ML Operations & Model Deployment

• Deploy and monitor ML models for soil carbon stock change prediction, biomass estimation, and ranch management photo verification systems
• Build inference pipelines that serve predictions to both internal dashboards and customer-facing applications
• Implement model versioning, A/B testing infrastructure, and rollback capabilities
• Optimize model performance for cost efficiency at scale (we’re running models across millions of acres)

Integration & Connectivity

• Build real-time data connectors for third-party hardware integrations (GPS tracking, virtual fencing, scale systems, weather stations)
• Develop APIs that expose ML predictions and aggregated data to internal teams and external partners
• Work with engineering to integrate data capabilities into Pasture Map’s mobile and web applications
• Support protocol compliance by ensuring data pipelines meet CAR, Verra, and emerging standard requirements

Collaboration & Technical Leadership

• Partner with Data Science to understand model requirements and translate research code into production-ready systems
• Work with Product to scope data requirements for new features and ensure technical feasibility
• Document systems thoroughly and establish operational runbooks
• Contribute to architectural decisions
  This role is roughly 50% data pipeline work, 30% model deployment/monitoring, 20% integrations