Fraud Data Scientist

at Provenir
Published January 9, 2023
Location England, United Kingdom
Category Data Science  
Job Type Full-time  


Who We Are

Provenir is a global fintech company with offices across North America, the UK, and Singapore. Provenir helps fintechs, financial institutions, and payment providers make smarter decisions, faster. We are passionate about technology and empowering businesses to become industry leaders. As a leading provider of decisioning and analytics products for financial services and other industries, we empower businesses to create digital-first decisioning solutions that drive business growth. If you’d like to work at an innovative fintech with a global footprint that is redefining the industry, then we want you!

This role can either be remote based, London based, Leeds based, or hybrid.


What You’ll Do

The Provenir AI team is a centralized function responsible for all things data, analytics, and ML/AI at Provenir. The team consists of Cloud Architects, Data Engineers, Software Engineers, Data Scientists, Product Mangers and Program Managers. It is our team's mission to disrupt longstanding players in the value chain by building a distributed AI/ML platform to drive best-in-class decisions.

We’re seeking a Fraud Data Scientist to apply state-of-the-art Machine Learning and Artificial Intelligence methods to large amounts of data from different sources to build and deploy modeling solutions in any one of the below domains (exposure to multiple domains is a bonus):

  • Anti-Money Laundering
  • Financial Crime
  • Transaction Monitoring
  • Anomaly Detection

In this role, you will create and validate AI and ML models that automate and enhance our fraud detection capabilities. Use cases could include models to predict: first-party fraud, third-party fraud, account take over, compromised ID detection, card not present, money laundering, social engineering.

Our ideal team member will have the mathematical, statistical and technology expertise you’d expect, but also a creative and curious mindset to help solve clients’ most demanding challenges.



  • Analyze data (past customer behavior, sales inputs, alternative data and other sources) to identify trends and create modeling solutions with clear recommendations
  • Determine indicators (patterns) of potential fraud and conceptualize ways to automatically detect instances. Identify relevant data sources to improve fraud and anomaly detection and compromised account prevention
  • Help identify and source relevant data (including alternative data) to build, calibrate and tune state-of-the-art risk models using the latest AI/ML technologies and approaches
  • Partner with Product and Engineering teams to support implementation in our cloud platform
  • Claim end-to-end ownership of your models and solutions: from ideation to implementation
  • While working in a fast-paced environment, interpret, document and successfully communicate analytic work and/or results to stakeholders, including those in non-analytic roles, generating recommendations to cross-functional teams and leadership
  • Be involved in the whole process of model development. This includes everything from root cause analysis, data collection and feature engineering to training, validating and implementing machine learning models, computing performance statistics and live model monitoring


What’s Required:


Bachelor’s degree in a highly numeric discipline, such as: statistics, mathematics, data science, engineering, physics, computer science, econometrics, actuarial science. Post-graduate degree such as Coursework Masters, Research Masters or PhD in related discipline is a bonus.


At least 5 years’ experience in data science role exposed to fraud and/or credit modeling

  • Direct hands-on experience in building end-to-end predictive models to classify fraud patterns such as: first-party fraud, third-party fraud, account take over, compromised ID detection, card not present, money laundering, social engineering
  • Development and monitoring of supervised and unsupervised advanced machine learning algorithms (gradient boosted decision trees, random forest, k-means clustering) using packages such as: scikit-learn, LightGBM, XGBoost, CatBoost, Keras.
  • Strong data wrangling and manipulation skills for merging, cleansing, and sampling data
  • Programming skills (at least one of): Python (pandas/numpy/seaborn), R (tidyverse), SQL
  • Comfortable working in a dynamic, research-oriented group with several concurrent projects
  • Experience with AWS is a bonus

Our employees are our top priority, we offer comprehensive health and wellness plans. You will enjoy paid time off and company holidays, flexible and remote-friendly opportunities, and maternity/paternity leave along with retirement benefits to plan for your future.


At Provenir, we recognize that diversity and inclusion make our teams stronger. We are committed to equal employment opportunity and welcome everyone regardless of race, color, ancestry, religion, national origin, age, sex, gender identity, sexual orientation, disability, marital status, domestic partner status, citizenship, or veteran status or medical condition. We encourage people from all backgrounds to apply.