Job Description & Details
This is a heavy-hitting senior contract role at a major financial institution focused entirely on building and tuning anti-money laundering models. If you enjoy digging into massive transactional datasets to catch financial crime and want to own the analytics side of risk scoring, this is a solid gig with potential for extension.
What You'll Actually Be Doing
Expect to spend a good chunk of your day writing code, tuning detection scenarios, and building behavioral models that keep financial crime out of the system. You will act as the analytics lead on transaction monitoring, which means you won't just be blindly executing tickets—you are expected to bring your own ideas to the table, challenge existing risk rules, and figure out better data-driven ways to flag suspicious activity without flooding compliance with false positives.
The Core Tech Stack
You need to be completely fluent in Python and PySpark, running on Databricks within an Azure environment, backed by heavy SQL querying. Because this team processes millions of transactions, your code needs to be optimized for scale; slow queries or poorly structured distributed dataframes won't cut it here. Experience with statistical modeling and predictive machine learning methods is non-negotiable since you'll be building the core risk-scoring engines.
Interview Expectations
Expect the hiring team to test your distributed computing chops with a tough question on optimizing a slow PySpark job that is bottlenecking a massive aggregation pipeline. They want to see if you understand shuffle partitions, caching strategies, and how memory management works under the hood in Databricks. You'll also likely get grilled on your practical approach to handling extreme class imbalance in fraud or AML datasets—specifically, how you evaluate models when the positive class (money laundering) is exceptionally rare.
Application Advice
Make sure your resume leans heavily into distributed data processing and production-level machine learning rather than just academic modeling. Explicitly highlight your hands-on experience with Python, PySpark, Azure, and Databricks so you easily clear the initial ATS keyword filters. If you have any past background in banking, risk scoring, or transaction monitoring, bring that front and center to prove you understand the domain.