Director Data Science
zohorecruit
Job Description
Core Responsibilities
Model Development
- Design, train, and evaluate machine learning models for production use
- Conduct experiments and A/B tests to validate model improvements
- Implement model interpretability and explainability techniques
- Stay current with latest research and apply state-of-the-art methods
Production ML
- Collaborate with Engineering and infrastructure to deploy models to production
- Build data pipelines and feature engineering workflows
- Monitor model performance and implement retraining strategies
- Create APIs and interfaces for model predictions
- Optimize models for latency, throughput and cost
Data Analysis and Insights
- Perform exploratory data analysis
- Identify patterns and anomalies in commercial real estate data
- Communicate findings to product and business stakeholders
- Develop metrics and dashboards to track model performance
- Validate data quality and implement data validation
Document Intelligence and NLP
- Build document extraction and classification models for loan documents
- Develop NLP pipelines for processing unstructured financial text
- Implement OCR and document parsing solutions for automated data extraction
Agentic AI and LLM Systems
- Design and implement LLM-powered applications and agentic workflows
- Develop RAG (Retrieval-Augmented Generation) systems for document Q&A
- Implement prompt engineering strategies and LLM evaluation frameworks
- Build guardrails and safety mechanisms for AI-generated outputs
Collaboration and Support
- Partner with Product teams to translate business requirements into ML solutions
- Work with Data Engineers on data pipeline and feature store requirements
- Collaborate with Domain Experts to validate model outputs against business logic
- Document model architectures, experiments, and decision rationale
- Mentor junior data scientists and share best practices