AI Researcher – World Models

Posted 8hrs ago

Employment Information

Education
Salary
Experience
Job Type

Report this job

Job expired or something wrong with this job?

Job Description

AI Researcher developing world models from geospatial and property data. Building forecasting capabilities for Geolava’s platform serving property and infrastructure decision-makers.

Responsibilities:

  • Design, train, and evaluate world models of the built environment that predict asset changes over time and in response to interventions and external forces
  • Develop representation learning and predictive modeling approaches over multi-temporal aerial and satellite imagery, property records, and other physical and administrative signals
  • Build rigorous evaluation methods using baselines, held-out temporal splits, calibration, and tests distinguishing genuine forecasting skill from leakage
  • Run experiments on cloud GPUs with attention to throughput and cost
  • Turn research results into clear, decision-ready writeups
  • Collaborate with AI and engineering teams to bring successful research into production in Geolava's platform
  • Stay current with world-model, self-supervised learning, and generative modeling literature and apply relevant ideas

Requirements:

  • PhD in machine learning, computer vision, or a closely related quantitative field
  • World-model research experience, including learned dynamics models, latent predictive models, action-conditioned generation, or closely related work
  • Demonstrated research work on world models or learned dynamics models, with publications, open-source work, or shipped systems
  • Expert-level PyTorch
  • Experience training and fine-tuning vision transformers or diffusion models at scale
  • Strong Python and data engineering skills for large image and tabular datasets
  • Experience running training workloads on cloud GPU infrastructure
  • Deep learning fluency, including training large vision models, adapting foundation models efficiently, and diagnosing training runs
  • Experience with remote sensing, geospatial data, or spatiotemporal modeling is a strong plus
  • Ability to design rigorous experiments with proper baselines, temporal splits, calibration, and leakage testing
  • Ability to work productively with ambiguous problems and small teams
  • Precise technical communication for research and non-research audiences

Benefits:

  • Competitive salary; compensation varies according to candidate level and experience
  • Medical, dental, and vision insurance
  • Retirement plan
  • Short- and long-term disability insurance
  • Life insurance
  • Global remote workforce