Financial Crime Product - Data Scientists & Engineers
Alexander Barnes are partnered with a high-growth fintech building out its financial crime product capability across fraud, transaction monitoring, and screening.
These roles sit within product and engineering, in the first line.
You’ll either be defining how detection works, or building the systems it runs on.
What you’ll be doing
- Building and improving detection systems across fraud and AML (card, banking, TM, screening)
- Working directly with transaction, auth, behavioural, and network data to identify patterns and signals
- Developing detection logic across rules, models, and AI
- Tuning systems continuously. False positives, detection coverage, operational load, customer impact
- Designing features and intelligence layers that improve how risk is detected
- Running deep analysis in SQL and Python. No reliance on dashboards
- Translating typologies into production-ready signals and decisioning logic
- Deploying and scaling models and rules in real-time systems
- Partnering closely with product, engineering, and compliance to evolve detection frameworks
Where this role can sit
Depending on your background, this leans into one of the following:
- Fraud Risk (Card / Banking)
- CNP, ATO, scams, APP, mule detection, onboarding abuse
- Working with auth data, payment flows, chargebacks, behavioural signals
- Transaction Monitoring (AML)
- Owning system performance. Rule effectiveness, typology coverage, backlog, false positives
- Working closely with ML models and monitoring frameworks
- Screening (Sanctions / PEP)
- Match quality, list coverage, tuning logic, global screening performance
- Detection Engineering / ML Systems
- Building monitoring frameworks from scratch
- Deploying models into production
- Scaling decisioning systems across large transaction volumes
What we’re looking for
- Hands-on experience in fraud, AML, or financial crime risk
- Strong understanding of typologies and how they translate into detection logic
- Strong SQL. Complex queries, large datasets, no hand-holding
- Python for analysis and modelling (pandas, numpy; ML exposure expected)
- Experience building, tuning, or deploying detection systems (rules and/or models)
- Ability to think about systems end-to-end. Not just models, but performance and outcomes
- Comfortable working across product, engineering, and compliance
Profiles that tend to work
- Fraud / risk data scientists from issuers, fintechs, or banks
- TM or screening specialists who understand system performance, not just policy
- Engineers who’ve built risk or monitoring systems at scale
- Investigators or law-side profiles who’ve moved into detection and pattern analysis
What doesn’t work
- Ops-only or case handling backgrounds
- Compliance or policy profiles without data or system ownership
- Engineers with no exposure to financial crime or risk systems
- People who can’t show what they’ve built, tuned, or improved
Why this role exists
Most teams measure financial crime after it happens.
This team is building the systems that detect it earlier, adapt faster, and scale properly.