AI Scientist – Geospatial & Multimodal Intelligence (US)

Pixxel is seeking a full-time AI Scientist specializing in Geospatial & Multimodal Intelligence. The role involves developing machine learning models using high-resolution hyperspectral satellite data for applications like crop classification and water quality retrieval. Candidates should have an advanced degree and 4+ years of relevant experience. Strong programming skills in Python and familiarity with geospatial tools are essential. The position offers flexible working hours and remote work options.
Department

Research & Modelling

Job posted on

May 30, 2025

Employment type

Full-Time

Role: AI Scientist – Geospatial & Multimodal Intelligence (US)
Employment Type: Full Time Eligibility: US Citizenship is mandatory
Educational Qualification:Advanced degree (M.S. or Ph.D.) in Computer Science, Remote Sensing, Geoinformatics, or a related field.

Work Experience: 4+ yrs of relevant experience.

Role Description:
We are seeking a Geospatial AI Scientist to join our Analytics team. This role focuses on developing state-of-the-art machine learning and deep learning models that harness Pixxel’s high-resolution hyperspectral satellite data and allied geospatial modalities. You will play a key role in creating scalable environmental intelligence tools by leveraging foundation models, multimodal fusion, and advanced AI techniques, and working closely with cross-functional teams.

Responsibilities & Duties:
• Design and implement geospatial AI models for applications such as crop classification, yield forecasting, water quality retrieval, and emissions detection.
• Contribute to the design and fine-tuning of large-scale geospatial foundation models for remote sensing tasks using self-supervised and transfer learning approaches.
• Develop multimodal architectures that integrate hyperspectral, SAR, weather, and ground-based data sources.
• Adapt models pretrained on well-labeled regions for use in label-scarce environments via transfer learning and domain adaptation.
• Incorporate scientific or physical priors into AI models (e.g., radiative transfer, crop phenology).
• Collaborate with data engineers and product managers to deploy models into scalable pipelines.
• Publish technical findings and contribute to thought leadership in geospatial AI.

Desirable Skills & Certifications:
• Proven experience in geospatial machine learning, including time-series analysis, land cover classification, or climate/environmental modeling.
• Strong programming skills in Python, with knowledge of ML libraries (PyTorch, TensorFlow, scikit-learn).
• Familiarity with geospatial foundation models (e.g., Prithvi, SatMAE, Segment Anything, ViTs trained on EO data).
• Experience with Earth observation tools and libraries such as Rasterio, xarray, PyProj, or STAC.
• Comfortable working with large-scale datasets and cloud-based geospatial processing.
• Experience with hyperspectral, multispectral, or SAR satellite imagery.
• Familiarity with cloud-based EO platforms like Google Earth Engine, SentinelHub, or OpenEO.
• Experience building or fine-tuning self-supervised or foundation models in a geospatial context.
• Demonstrated ability to fuse EO imagery with climate, terrain, or in-situ sensor data.
• Track record of publications or contributions to the remote sensing or AI research community.

Candidate Acumen:
The candidate should have strong expertise in deep learning and geospatial analytics, with hands-on experience in building scalable AI models using satellite or hyperspectral data. Proven ability to work on foundation models, multimodal fusion, and environmental intelligence applications is highly desirable.

Benefits:
Health insurance coverage
Unlimited leaves & flexible working hours
Role-based remote work and work-from-home benefit
Relocation assistance
Professional Mental Wellness services
Creche facility for primary caregivers (limited to India)
Employee Stock Options for all hires

About Pixxel

Pixxel is a space data company, building a constellation of hyperspectral earth imaging satellites and the analytical tools to mine insights from that data. Pixxel’s constellation is designed to provide global coverage every 24 hours and help detect, monitor, and predict global phenomena across agriculture, mining, environment and energy use cases.

Pixxel is on a mission to build a health monitor for the planet through a constellation of cutting-edge hyperspectral small satellites. This unique hyperspectral capability will be the key to unearth underlying, unseen problems that are invisible to satellites in orbit today.

Pixxel has collaborated with prominent organizations such as the Indian Space Research Organization, NASA JPL, and SpaceX among other leaders in the space industry. The organization is backed by Lightspeed, Radical Ventures, Relativity's Jordan Noone, Seraphim Capital, Ryan Johnson, Blume Ventures, Sparta LLC and Accenture among others.

Recently, Pixxel successfully raised $36 Million in its Series B funding round, with participation from new investors like Google, alongside existing investors, bringing the company's total venture funding to $71 Million.

To know more about us, visit https://www.pixxel.space. :
https://www.notion.so/pixxelspacetech/Pixxel-in-Media-92a5072f085e4300a0defa1ba2d37149