SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Forty5Park is building next-generation AI systems for real estate, leveraging data science, intelligent automation, and advanced machine learning to uncover insights and drive smarter decisions across the industry. The team combines rigorous experimentation with scalable technology, using data-driven methods, generative AI, and mathematical modeling to transform how the built environment is understood and optimized.
You'll join the Applied AI Research team as a Machine Learning Researcher, applying a scientific, hypothesis-driven approach to build models that forecast market trends, asset valuations, and other key real estate dynamics. You'll work at the intersection of machine learning, econometrics, and applied statistics, exploring innovative modeling techniques from Bayesian inference and Markov processes to deep learning architectures for time-series forecasting.
Key responsibilities include: designing, developing, and testing forecasting models for property valuations, price trends, and market risk; applying machine learning, Bayesian inference, and Markovian modeling to large-scale multidimensional datasets; formulating hypotheses, designing experiments, and analyzing results using rigorous statistical methods; partnering with engineers to prototype, validate, and deploy predictive models into production; contributing to research culture through reading groups and seminars; staying current with advances in time-series forecasting and causal modeling; and communicating insights to technical and business stakeholders.
Required qualifications: Advanced degree (Master's or PhD) in Statistics, Mathematics, Computer Science, Physics, or Engineering (exceptional candidates with strong applied research experience considered); deep understanding of machine learning algorithms, statistical inference, and forecasting techniques; demonstrated experience with real-world datasets, preferably in econometrics, geospatial analysis, or financial modeling; proficiency in Python or similar languages; strong background in probabilistic modeling including Bayesian methods, Markov processes, or Monte Carlo simulations; proven ability to perform independent research and translate results into actionable solutions.
Preferred experience includes familiarity with deep learning frameworks (PyTorch, TensorFlow), modern forecasting architectures (Transformers, Temporal Fusion Networks), spatiotemporal data modeling, graph-based learning, causal inference techniques, academic publications or conference presentations, and understanding of real estate data systems or investment modeling.
The role is full-time, onsite in Manhattan (5 days per week), with unlimited PTO, dog-friendly office, continuous learning support including conference sponsorships, and opportunity to work with industry-leading AI talent on high-impact problems.
