Head Of Data Science

About ClearGrid

ClearGrid is on a mission to revolutionize the debt resolution industry. We’re using AI, automation, and real-time data to completely rethink how debt collection works, unlocking better outcomes for institutions and individuals alike. Think of us as rebuilding the collections —starting from scratch, with modern tools.

Were a fast-growing startup with ambitious goals, and we’re looking for builders who are hungry to make a real dent in a space with massive untapped potential.

We’re looking for a Head of Data Science & Intelligence - a technical leader who can turn our massive data ecosystem into predictive systems that power everything from customer behavior modeling to portfolio valuation and NPL purchase strategy. This isn’t about dashboards. It’s about building the brains of a $500B+ AI-driven financial platform.

Your Mission

You’ll architect, build, and deploy the models and data systems that make ClearGrid self-learning. From repayment probability to portfolio pricing, from conversation intelligence to cashflow forecasting — your models will be the decision engine behind how capital moves through the ecosystem.

What You’ll Do

• Architect and build predictive models across the stack — customer segmentation, PTP likelihood, loss-given-default, recovery rate prediction, and portfolio valuation.
• Operationalize ML pipelines using Python, SQL, and MLOps frameworks (Airflow, dbt, TensorFlow, PyTorch, or equivalent).
• Design feature stores and data schemas that enable real-time intelligence for AI voice agents, dashboards, and risk engines.
• Build and fine-tune models for securitization and NPL pricing, integrating macroeconomic variables, borrower behavior, and cashflow projections.
• Integrate AI/ML into business operations — connecting model outputs directly into CRM, call orchestration, and workflow automation.
• Collaborate with AI engineers to align model input/output structures with LLM systems and conversation intelligence modules.
• Own the full model lifecycle — data prep, model design, validation, deployment, monitoring, and retraining based on human-AI feedback loops.
• Develop behavioral segmentation frameworks that dynamically group borrowers by risk, responsiveness, and emotional tone to optimize multi-channel strategy.
• Create financial intelligence systems for loan book pricing, debt sale structuring, and securitization analytics — bridging data science with finance.
• Mentor analysts and data engineers, establishing coding standards, version control, model documentation, and review processes.
  
  

• 6–10+ years in data science or quantitative modeling, ideally in fintech, credit, or analytics-heavy startups.
• Expert in Python, SQL, and statistical modeling — you code, experiment, and debug at production level.
• Deep understanding of machine learning, probabilistic modeling, NLP, and predictive analytics.
• Experience deploying ML models to production (e.g. MLflow, Vertex AI, SageMaker, or custom pipelines).
• Familiarity with financial modeling and credit analytics, including portfolio risk, pricing, and securitization concepts.
• Hands-on with modern data architecture — warehouses (BigQuery, Snowflake), orchestration (Airflow, Prefect), and data versioning (dbt, Git).
• Proven ability to transform raw, messy data into structured insights and usable systems.
• Experience building and managing data feedback loops between human agents, AI models, and product decisions.
• Comfort designing end-to-end data systems that serve both predictive modeling and operational analytics.
• Bonus: exposure to LLM integration, embeddings, and model alignment methods.
  
  

• A builder-scientist hybrid — you think like a researcher but deliver like an engineer.
• Obsessed with systemic intelligence — you see how every datapoint connects to operations, behavior, and capital markets.
• Relentlessly curious and allergic to stagnation — if there’s a better model, tool, or method, you’ll find it.
• Fluent in both business logic and mathematical rigor — you can explain a model to a CFO and a neural net to an engineer.
• The rare ability to translate between business strategy and data systems — turning messy, cross-functional chaos into measurable advantage.
  
  

• You’re not a BI dashboard builder.
• You’re not a report generator.
• You’re not waiting for business teams to tell you what to analyze.