Data Scientist - Satellite Imagery & AI Agents

About the role

Planet is looking for a Data Scientist to build validated markers that extract insights from dense temporal stacks of satellite imagery, covering agriculture, land management and climate. The markers form the core of the Area Monitoring System (AMS) delivered to Common Agricultural Policy paying agencies across Europe, and the role involves deploying models at scale with scientists and engineers, moving beyond compliance into land-cover change detection with embeddings and AI-first workflows. It is a full-time hybrid role, working from the Graz, Ljubljana or Haarlem office three days per week.

Key responsibilities

• Develop algorithms and machine learning models that extract insights from satellite imagery time series, and maintain and improve existing markers.
• Take a research topic from exploration through to production.
• Own marker results for production AMS deliveries across several EU countries, running them, reviewing them and confirming quality before delivery to the client.
• Explore and build embeddings-based insight extraction in new areas beyond CAP.
• Use AI agents in daily work and help the team get better at them.
• Co-own the markers codebase and collaborate on iterative solutions on the platform and existing data building blocks.
• Document work so it is transparent and repeatable, and write internal research reports, public blog posts and client reports.

Required profile

• 4+ years of relevant work experience.
• Bachelor's degree or higher in computer science, data science or another STEM field.
• Understanding of machine learning principles, including model validation and performance evaluation.
• Working knowledge of the geospatial domain; English at a professional level.
• Ability to deliver projects on schedule and manage technical deliverables.

Required skills

• Proficiency in Python, with maintainable, well-documented code.
• Experience with AI agents and modern machine learning tools.
• Version control and Git, and comfort in a shared codebase.
• Nice to have: embeddings or learned image representations, remote sensing and satellite time series, radar data processing (e.g. Sentinel-1), cloud environments and distributed computing, and familiarity with the agricultural domain or EU CAP.

What we offer

• Paid time off including vacation, holidays and company-wide days off.
• Employee wellness program, home office reimbursement and monthly phone and internet reimbursement.
• Tuition reimbursement and access to LinkedIn Learning.
• Equity and possible discretionary bonuses.