Senior ML Engineer — Credit Risk & Compliance

Anton Volkov

Full-time · Senior · Toronto

This role is no longer accepting applications

Similar open roles you can apply to now:

About the role

We are a Canadian financial technology company regulated by FINTRAC and operating under OSFI supervisory guidelines. We provide lending products to small and medium-sized businesses across Canada — working capital facilities, equipment financing, and trade credit. Our loan book is CAD $1.8 billion. Our credit decision models are our core intellectual property. They are also subject to regulatory scrutiny. Every model we use in a credit decision must be explainable, must be tested for discriminatory impact, and must be documented in a form that satisfies an OSFI model risk management review. We have a credit analytics team of twelve people — credit scientists, model validators, and data engineers. We are hiring a Senior ML Engineer who can build and maintain credit risk models at the level of technical sophistication our book requires, while operating within the model governance framework that our regulatory obligations demand. This is not a role for someone who wants to build models without thinking about what they will be used for. It is a role for someone who finds the constraint of building explainable, auditable ML in a regulated domain technically interesting rather than frustrating.

Responsibilities

Design and build credit risk models for underwriting, pricing, and portfolio monitoring across our SME lending products
Own model validation documentation: performance monitoring reports, champion/challenger analysis, and model risk assessments for our internal validation function
Collaborate with the credit policy and compliance teams to ensure models satisfy fair lending requirements and regulatory expectations
Contribute to the model governance framework: model inventory, performance trigger definitions, and the escalation process for model risk events
Mentor junior credit scientists and data analysts on model development standards and regulatory documentation requirements

Requirements

7+ years of ML engineering with at least 3 years building credit risk or financial risk models in a regulated environment
Python at a senior level: production-quality modelling code, clean APIs, and the discipline to write model implementations that a validator can audit
Scikit-learn for credit scoring, survival analysis, and the classical approaches that remain best-in-class for explainable lending decisions
SQL for working with our loan origination data, bureau data feeds, and performance tracking datasets at scale
Statistics at depth: logistic regression, calibration, Gini coefficient, KS statistic, PSI — the metrics of credit risk model performance that a regulatory submission requires
Snowflake for our data warehouse — you should be comfortable with large-scale SQL and understand how to write queries that a data engineer can maintain
Familiarity with OSFI E-23 Model Risk Management guidelines or equivalent regulatory model governance frameworks is a significant advantage