Data Scientist

Findigs•New York, NY

8d•$160,000 - $185,000•Hybrid

About The Position

Findigs runs an AI underwriting engine (DecisionAssist) that makes or influences thousands of rental decisions every week. As Data Scientist at Findigs, you will strengthen our data science and applied machine learning depth: owning hands-on model development, experimentation design, and ML-adjacent analysis that directly impacts renter and property manager outcomes. Reporting to the Lead Analytics Engineer, this is a highly technical, high-ownership role for a data scientist who wants to build and improve production models, bring statistical rigor to product decisions, and grow into broader strategic scope as the team evolves. You will partner closely with Product and Engineering to translate real-world rental risk and behavior into models, experiments, and clear insights. Please note, we are unable to sponsor or take over sponsorship of an employment visa at this time.

Requirements

4+ years of hands-on data science or applied ML experience (fintech, proptech, or other high-stakes decisioning environments preferred)
Strong Python skills (pandas, scikit-learn, statsmodels or equivalent); this is a coding role
Ability to design, run, and interpret A/B tests independently
Strong SQL skills and comfort working in a modern data stack (dbt, Snowflake, Sigma, or similar)
Solid grounding in supervised learning fundamentals (classification, regression, tree-based methods)
Strong written communication and the ability to explain model behavior and tradeoffs to non-technical partners (e.g., PMs, CSMs)
Intellectual curiosity about housing and credit data in particular

Nice To Haves

Experience building or contributing to a credit, risk, or underwriting model in production
Familiarity with fair lending / disparate impact considerations in ML (important given the real-world consequences of renter screening)
Experience working on systems where model output directly affects real people, with a strong sense of responsibility and rigor
Ability to move between exploratory research and production-grade work without needing separate tracks
LLM experience (fine-tuning, retrieval, or integration), especially as we automate parts of underwriting and screening workflows
Startup / scale-up experience

Responsibilities

DecisionAssist model development: Own feature engineering, model iteration, and evaluation for DecisionAssist. You will work across two surfaces: (1) operational model work in the DA/CAV1 serving layer, and (2) analytics-focused modeling in Snowflake for experimentation and research, as well as partner with Product and Engineering on what signals matter and why.
Experimentation and A/B testing: Design and analyze experiments across underwriting, renter-facing, and PMC-facing product changes, and bring statistical rigor and clear recommendations.
Predictive and risk modeling: Build and maintain models used in screening logic (e.g., delinquency risk, income estimation, fraud signals).
ML infrastructure: While you won’t own the warehouse or pipeline architecture, you should be comfortable writing clean Python, working in dbt, and operating in a modern data stack.
Research and analysis: Tackle high-impact, ad-hoc questions from Product and Customer teams; e.g., what’s driving approval-rate variance, which cohorts behave differently, and what a given signal actually predicts.