Geospatial Machine Learning Engineer

Job Title: Geospatial Machine Learning Engineer

Location: Bangalore, India

Experience: > 2 years of experience in Machine Learning, Data Engineering, Geospatial Analytics, or related fields, with hands-on exposure to building and deploying ML systems using large-scale geospatial datasets.

About Alt Carbon:

Alt Carbon is a deeptech science and data company, building agri-infrastructure for climate action. We aim to make South Asia a hub for Carbon Dioxide Removal (CDR) through technology pathways like Enhanced Rock Weathering and Biochar. We work with farmers and scientists across the Global South to turn underutilized land into carbon sinks.

Our flagship initiative, the Darjeeling Revival Project (DRP), is a first-of-its-kind effort to unite climate action with cultural and ecological restoration - by reviving degraded soils, restoring livelihoods, and rebuilding ecosystems. This year, we also launched the Bengal Renaissance Project (BRP), a large-scale biochar program that is building a network of biomass banks, pyrolysis, and gasification plants across Eastern India, with a mission to weave hope back into the lands of India.

Rooted in science, powered by community, and driven by the belief that revivals require ambition and audacious bets, our mission is to remove 5 million metric tons of CO₂ by 2030.

About the Team:

Every ML practitioner's first dataset was Iris — a small set of measurements that taught the world to classify what it could see. Our Iris team carries that spirit forward, only the dataset is the planet. We are Alt Carbon's engineering and AI backbone, building the foundational data infrastructure and models to read the earth's soil, atmosphere, and geology at scale. The work is early, the ambition is not — we're laying the groundwork for planetary intelligence, one pipeline at a time.

Role Overview:

Alt Carbon is building the infrastructure for Planetary Intelligence — turning fragmented observations about soil, water, atmosphere, and rock into a unified, actionable understanding of Earth. We are looking for a Geospatial ML Engineer to make this real at the data and model layer: fusing multi-modal datasets across remote sensing, geochemistry, and climate systems into scalable pipelines, and building robust, interpretable ML models that work in production, not just in notebooks. If you get a thrill from turning a research paper into running code, we'd like to talk!

Responsibilities:

Data Pipelining

• Build a scalable data pipeline for publicly available (e.g., NASA, ESA, USGS, ISRO, FAO, Copernicus) and proprietary multispectral, hyperspectral, atmospheric, and soil/geology datasets.

Model Engineering

• Work alongside data scientists and subject-matter experts to harden prototype ML models for production, leveraging multiple data modalities (satellite imagery, soil chemistry, weather patterns).
• Build and maintain experiment tracking and model registries so every model version is traceable from training data to deployment.

Deployment & Scaling

• Optimise and deploy models for large-scale inference on cloud/on-prem infrastructure.
• Implement monitoring for data drift and model drift, with automated alerting.

Requirements

• B.Tech/M.Tech in Machine Learning, Data Science, Computer Science, Geoinformatics etc. with 2-3 years experience as a Data Engineer/ML Engineer
• Hands-on experience with processing geospatial data (Sentinel, Landsat, MODIS, ERA5, etc.) and formats (GeoTIFF, NetCDF, ESRI Shapefiles)
• Proficiency in Python, with experience in ML/DL frameworks (PyTorch, scikit-learn, XGBoost, LightGBM).
• Familiarity with geospatial processing libraries (GDAL, rasterio, geopandas, xarray, rioxarray).
• Experience deploying models on cloud like AWS (Sagemaker, S3, EC2)
• Experience with ML lifecycle tooling (like MLflow) and workflow orchestration (like Airflow).

Nice to Haves

• Familiarity with cloud-native geospatial stacks (e.g., STAC catalogs, cloud-optimized formats).
• Contribution to open source ML projects.
• Understanding of soil science, climate modeling, or geology is a plus.

What We Offer:

• A front-row seat to building Planetary Intelligence from the ground up — with access to one of the richest multi-modal earth science datasets being assembled anywhere, spanning satellite imagery, drone captures, sensor networks, lab spectrometry, and field measurements, all under one roof.
• Shape the tooling and practices of a young engineering team — your architectural decisions will set the foundation, not inherit someone else's.
• The velocity of a 50-person startup with the scientific depth of a research lab — your code goes from PR to production to real decisions, fast.

Why Join Us?

Every machine learning engineer dreams of working on a problem that actually matters. Few get the chance.

At Alt Carbon, your models won't be predicting what someone clicks next. They'll help us understand soil, rock, weather, and ecosystems at planetary scale. You'll work with satellite imagery, field data, laboratory measurements, and one of the most unique earth science datasets being assembled anywhere.

The challenge is hard. The data is messy. The stakes are real.

If you're excited by the idea of building machine learning systems that help humanity better understand the planet we live on, this might be the most meaningful work you'll do.