Applied Research Scientist – World Model

This is a fully remote position, open to applicants in United States.

📋 Description

• Develop the demand layer of Zillow's World Model utilizing extensive shopper behavior data.

• Transform search, engagement, and session pathways into organized demand representations, including explicit attention-to-demand conversion.

• Design and train latent state-space, multi-tower, and/or sequential dynamics models with calibrated uncertainty.

• Facilitate multi-step rollouts, forward planning, and backward planning.

• Assess willingness-to-pay, per-home demand curves, and market-clearing dynamics.

• Link learned representations to economic metrics.

• Implement causal identification techniques on observational data.

• Specify when the model should refrain from making “what-if” claims when correlations are not supportive.

• Contribute to the evaluation framework of the World Model, which includes state probes, calibrated multi-target forecasts, rollout drift, causal checks, and out-of-regime replays.

• Independently lead ambiguous research projects from start to finish.

• Offer technical guidance and mentorship while converting cutting-edge research into practical applications.

• Collaborate with the World Model working group along with product, engineering, platform, and data teams.

⛳️ Requirements

• PhD in Computer Science, Machine Learning, Statistics, Economics, or a related quantitative discipline.

• Practical experience in developing latent state-space or world models, including RSSM/Dreamer-family, JEPA-style latent prediction, or event-sequence transformers.

• Familiarity with belief-state representations, stochastic transition models with calibrated uncertainty, and multi-step rollout evaluation.

• Capability to convert extensive user behavior data into structured demand representations, including utility or willingness-to-pay estimation and funnel/conversion modeling.

• Background in correcting for selection bias in platform-observed populations.

• Practical experience in causal inference within observational contexts, including quasi-experiments, instruments, and policy discontinuities.

• Discerning judgment on when learned correlations justify counterfactual assertions.

• Documented success in delivering open-ended research initiatives.

• Aptitude for defining ambiguous issues and providing technical guidance to scientists and MLEs.

• Proficient programming skills, particularly in Python.

• Experience with machine learning frameworks such as PyTorch or TensorFlow.

• Strong economics knowledge in market microstructure, search-and-matching, or structural/discrete-choice demand models is highly desired.

• Experience with agent-based or market simulation, or simulation-based reinforcement learning.

• Published research in world models, model-based RL, causal inference, or structural demand estimation.

• Familiarity with multimodal representation learning or bitemporal, leakage-safe training data.

• Awareness of two-sided marketplaces, real estate, or long-term consumer journeys.

🏝️ Benefits

• Equity awards determined by factors such as experience, performance, and location.

• Option for remote work from a location of the employee's choice.

• Flexibility to work from wherever is most productive.

• Support for accommodations regarding disabilities or special needs.