Senior Data Scientist, Research & Applied Science
Location: Hoboken, NJ (on site)
About Faropoint
Faropoint owns and operates last-mile, Class B, small-bay industrial properties across 20+ U.S. markets. Our in-house technology team builds proprietary software and predictive models that shape what we buy, how we price leases, and which tenants we underwrite. It's a senior group that works side by side with the investors and asset managers who use its tools every day.
The Role
Our models estimate market rent, forecast lease renewals, price property sales, and score tenant credit risk. Their outputs feed multimillion-dollar decisions. Small-bay industrial is fragmented, comps are thin, and no two assets are alike. This is a judgment problem, not a big-data problem.
You'll be our senior data scientist: the person who can authoritatively explain how our models work, including to an investor's own data scientists, and who decides whether they're right and proves it. You'll set our standards for validation and experimentation, work with the Director of Research on what to test next, and mentor junior data scientists who help maintain the models. No real estate background is required.
What You'll Do
Own the core prediction models
- Retrain, improve, and extend our rent, lease renewal, sale pricing, and tenant credit models, plus entity de-duplication.
- Pick up the record of hundreds of past experiments, including the ones that failed, and turn it into a research baseline so we don't repeat them.
- Benchmark new approaches, including tabular foundation models, AutoML, and agent-driven automated research, against well-tuned gradient-boosted baselines. Adopt them only when they win on held-out, real-world outcomes.
- Test candidate data sources and features (drive time, location quality, zoning, site characteristics, our spatial layers) against actual leasing and sales outcomes, and decide what earns a place in the models.
Make the outputs trustworthy
- Build backtesting, error analysis, and monitoring so model outputs are verified before they reach decision-makers.
- Design experiments that stay honest at scale. Sealed holdouts, time- and location-aware splits, and multiple-comparison controls mean hundreds of automated experiments still produce conclusions you'd defend.
- Keep research reproducible: versioned training data, documented features, repeatable experiments. Document the models as they stand today, since the existing documentation is out of date.
Run models in production
- Own serving endpoints, drift monitoring, and a model registry with clear lineage. Make inference reliable and fast (batch prediction runs and SageMaker endpoints that don't fail under load) and keep serving costs in line with actual usage. You'll work on the AWS platform and deployment pipelines the platform engineer provides.
Evaluate our LLM pipelines
- Own the methodology for measuring how accurately LLMs extract data from leases and market reports: labeled datasets, calibrated automated graders, and a clear definition of what counts as a regression.
Partner with the business
- Turn questions from investment, research, and asset-management colleagues into analyses, and explain uncertainty in plain language.
- Represent our models externally when an investor or partner wants to understand them.
- Track developments in ML and AI, and bring back what proves useful on our problems.
What You'll Bring
Must-haves
- 6+ years of applied statistics or machine learning, including production models you kept owning after launch.
- Strong Python and SQL, and deep experience building predictive models on tabular data.
- Rigor in validation and experimentation: you know how a model can look better than it is, and you can explain error and uncertainty to a non-technical audience.
- Experience deploying and monitoring models in production on a cloud platform (AWS SageMaker or equivalent).
- Daily use of AI coding assistants (Claude Code, Cursor, or similar), with the judgment to check what they produce.
Nice to have
- A Master's or PhD in Statistics, Computer Science, Economics, Operations Research, or a related field.
- Experience modeling real estate, credit, or asset prices (rent or valuation models, survival or renewal models, credit risk).
- Spatial data experience (GeoPandas, PostGIS, H3, ArcGIS, drive-time or routing analysis).
- MLOps experience: faster inference, stable endpoints, cost-efficient serving.
- Experience building evaluations for LLM-based document extraction.
- Hands-on experience with tabular foundation models, AutoML, or agent-assisted feature and hyperparameter search, including where they fail.
- Familiarity with the AWS data stack (Athena, S3, Glue, dbt, Redshift).
- A background in causal inference or econometrics.
What We Offer
This position offers a base salary in the $160,000-$200,000 range, plus discretionary bonus, competitive healthcare benefits, unlimited PTO, and a 401k match.
Equal Opportunity Employer
Faropoint is proud to be an Equal Opportunity Employer. We value diversity, and we do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, parental status, disability status, or any other status protected by local law. We believe that our work is better and our company culture is improved when we encourage, support, and respect the different skills and experiences represented within our workforce.
AI Notice to Candidates. Faropoint Ventures LLC and its subsidiaries (the “Company”) and their recruiting vendors may use artificial intelligence and other automated tools (“AI Tools”) to parse, screen, rank, assess, and summarize resumes/CVs, applications, and interviews. No hiring or rejection decision will be made solely by automated means without meaningful human review by Company personnel, and any use of AI Tools will comply with applicable law, including anti-discrimination laws and laws governing automated employment decision tools and algorithmic decision-making. Personal information you submit will be kept confidential in accordance with our Privacy Policy and used for recruiting, hiring, and related legal, compliance, and recordkeeping purposes, including to evaluate you for employment opportunities with the Company and its affiliates.