Lead Risk Strategist, AI & Payment Fraud (Relocation to Toronto Required)

Snaplii•New York, NY

22h•Hybrid

About The Position

At Snaplii, risk management is viewed as a growth enabler rather than a constraint. The company is experiencing rapid expansion and seeks a strategist who can leverage AI to identify and mitigate fraud risks. This role is crucial for defending a multi-million dollar ecosystem processing over $100M in annual transaction volume (TPV) and serving more than 250,000 users. The position offers the opportunity to work in a high-velocity environment, integrating AI models for automated fraud interception and building user behavior scoring and anomaly detection systems. It also provides exposure to advanced anti-fraud strategies through connections with payment processors and AI leaders.

Requirements

Minimum of 3 years dedicated to fraud and at least 1 year within the payments industry.
Strong problem-solving and reverse-engineering skills, with the ability to anticipate risks from a fraudster's perspective.
Proven ability to identify fraudulent activities across various payment methods in multi-currency, e-commerce environments.
Mastery of SQL for independent querying of large datasets.
Python proficiency is highly preferred.
3+ years of experience building fraud detection models, including feature engineering, model training, and evaluation.
Skilled in Python/R frameworks such as Sklearn, XGBoost, and LightGBM with experience deploying models into production.

Nice To Haves

Relocation to Toronto, Canada is required.
Onsite presence in Toronto for at least the first 6-12 months is required.
Experience with Hacker Mindset and raw analytical horsepower.

Responsibilities

Lead end-to-end financial risk strategies, including identification, design, testing, and post-production monitoring.
Investigate anomalous activity and perform root-cause analysis on chargebacks.
Produce authoritative reports on fraud trends.
Act as a liaison between Snaplii and payment processors/vendors, managing relationships and ensuring alignment on risk management.
Integrate AI models into risk systems to achieve high rates of automated interception of fraudulent behavior.
Build user behavior scoring and anomaly detection systems using Machine Learning (XGBoost, LightGBM).