Senior AI & ML Engineer - AML Search Quality

Description

The company provides a platform for fraud prevention and AML compliance. It serves thousands of companies worldwide, enriching customer profiles from more than 900 real-time, first-party data signals, flagging suspicious behavior and simplifying compliance workflows in one place. Customers include well-known digital brands such as Revolut, Wise, and Bilt, and the company reports fraud reductions of up to 95% and a 32x ROI for clients.

The company is building a new AML Search Quality team to make AML name screening more accurate and to grow its data science work in this area. The team combines data scientists, who improve name matching across sanctions, PEP, and adverse media data through fuzzy search scoring, configuration tuning, and match quality evaluation, with engineers who build and run the fuzzy search services that apply those models to customer screening in production.

The goal is fewer missed true matches and fewer false positives, measured in a way the team can track. You will own modeling and experimentation for the data science work, from defining the problem through to production, and work closely with engineers. You will also help set the team's technical direction, in a domain where accuracy affects customers' compliance outcomes.

The role can be based remotely within the EU.

Responsibilities

• Lead modeling and experiments for fuzzy matching and AML screening quality, turning unclear problems into clear plans and taking them from experiment through release and ongoing operation
• Create evaluation datasets and metrics from offline and online signals to measure match quality and guide changes to models and configurations
• Design scoring and ML methods that work within real-time decision limits, weighing match quality against speed, cost, and operational effort
• Take responsibility for production quality of your models and configurations, including drift monitoring, on-call duty, and follow-up after incidents
• Look into customer-reported matching problems with Product and Customer Success, and make sure customer feedback shapes priorities
• Work with Product, QA, Architects, DevOps, and Customer Success
• Record what experiments teach so the team can reuse it, and bring in AI tools and techniques where they clearly help
• Mentor teammates and push the team's data science practice forward

Requirements

• 5+ years of experience in data science or machine learning, with models delivered to production
• Strong Python skills
• Strong grasp of statistics and experiment design, including offline and online evaluation of search or matching quality, such as trade-offs between precision and recall
• Experience with NLP or text-based machine learning, such as text classification, named entity recognition (NER), or text matching
• Experience owning the full lifecycle of data-driven systems, from data and experiments through deployment and monitoring, while anticipating data drift, leakage, and bias
• Solid background in data analysis and SQL, with a working understanding of backend services and APIs
• Experience with non-relational or search databases, ideally Elasticsearch
• Hands-on experience with a major cloud provider, ideally AWS, including deploying models and data pipelines
• Clear communication, so you can explain modeling trade-offs to Product and non-technical stakeholders
• A record of turning unclear problems into end-to-end solutions and flagging delivery risks early along with a proposed mitigation
• Nice to have: prior experience in AML, KYC, or another regulated financial services domain
• Nice to have: experience leading cross-team technical work
• Nice to have: experience using AI-assisted development tools and helping a team apply them with good judgment

Conditions

Remote work within the EU