Qualifications

• Strong experience with graph machine learning techniques, including representation learning and graph neural networks on large-scale datasets.
• Expertise in entity resolution and record linkage, focusing on maintaining cluster stability over time.
• Familiarity with community and cluster detection using both classical and embedding-based methodologies.
• Proven success in identifying and analyzing coordinated or collusive behavior, particularly in fraud or AML contexts.
• Practical experience in building scalable production machine learning pipelines, especially for graph data.
• Exposure to financial crime, market surveillance, or trading environments, enhancing relevance to the role.
• Knowledge of FIX protocol and market microstructure is advantageous; experience with on-chain and EVM primitives is highly preferred.

Responsibilities

• Build and maintain the entity graph that associates accounts, wallets, and behaviors to identify actual actors.
• Develop enduring entity-resolution and clustering methods to reliably recognize returning actors.
• Detect and analyze coordinated behaviors and collusion rings, including situations without explicit identity sharing.
• Collaborate on the Attribution Substrate by integrating vendor attribution with on-chain behavioral clusters for a robust entity-resolution layer.
• Construct graph-based features and representations that support the broader detection system.
• Partner with engineering teams to expand the graph with new data sources and market types.

Benefits

• Onsite position based in Santa Monica, CA or New York, NY offices.
• Full range of medical, financial, and other benefits offered.