Description

You will translate and modernize complex quantitative models into high-performance, scalable ML systems.

This role is on-site.

Responsibilities

• Reverse-engineer legacy C++ and R codebases to extract core mathematical logic and simulation parameters.
• Re-implement extracted models in PyTorch, using torch.nn and custom Autograd functions.
• Refactor code to leverage Databricks distributed computing and PyTorch GPU capabilities for execution time reduction.
• Build high-performance data pipelines from Snowflake into Databricks using Spark and PyTorch DataLoaders.
• Conduct rigorous back-testing and sensitivity analysis to validate PyTorch model parity against legacy Hadoop outputs.

Required Skills

• 8+ years of experience in quantitative modeling or ML engineering.
• Expert-level proficiency in PyTorch for time-series, regression, or Monte Carlo simulations.
• High proficiency in Python, with strong ability to read C++ and R (including statistical packages like lme4 or forecast).
• Hands-on experience with Databricks (MLflow, Spark) and Snowflake.
• Deep understanding of statistical modeling, econometric forecasting, or financial risk management.
• Experience migrating workloads from Hadoop/Hive environments.

Preferred Skills

• Experience with PPNR, CCAR, or DFAST regulatory modeling.
• Familiarity with TorchScript or ONNX for model productionization.

Key Skills

Education

Any Gradute