Data Scientist

Responsibilities

• Design, train, and maintain statistical models (scoring, valuation, anomaly detection, segmentation)
• Rigorously assess the quality of models and data — including data provided by external vendors
• Leverage geospatial data (POIs, flows, zoning, distances, cadastral data and DVF — geolocated property transaction data) to enrich valuation and scoring models
• Produce actionable outputs for product and business teams
• Contribute to the data roadmap by identifying high-value opportunities
• Design and maintain training and inference pipelines
• Work closely with the Data Engineer on data architecture and input data quality
• Participate in model orchestration and production monitoring
• Integrate LLM / RAG / generative AI approaches into product use cases when appropriate
• Design and deploy RAG pipelines over real estate and regulatory document corpora
• Master AI engineering patterns: structured prompt engineering, function calling, lightweight AI agents, orchestration chains (Lang

  Graph, Crew

  AI)

Requirements

• Strong proficiency in Python for data science (Pandas, scikit-learn, statsmodels, Num

  Py, Geo

  Pandas, Shapely)
• Hands-on experience with supervised and unsupervised modeling on structured data
• Ability to rigorously assess external data (quality, bias, drift)
• Ability to work autonomously in a fast-evolving/startup environment — no need for a highly formalized framework to move projects forward
• Exposure to real estate, financial, or market data — a real plus
• Familiarity with generative AI and practical use of current tools (LLMs, RAG, assisted automation)
• Ability to communicate complex results to non-technical stakeholders