Warszawa, Warszawa, Polska
PAYBACK
Pełny etat
Role OverviewWe are looking for a hands-on Data Scientist to support our fraud recognition project.
The role owns the data science and machine learning aspects of the solution, from data analysis to model evaluation.
The role is well suited for colleagues with strong analytical, SQL, Python, or software backgrounds who want to grow into a machine‑learning‑focused position.
Your
responsibilities:
Analyze transactional and behavioral data to identify fraud patterns; perform data quality checks and exploratory analysisPrepare and document datasets for training, validation, and testing using SQL and PythonDesign fraud-relevant features and translate business logic into measurable signalsDevelop, evaluate, and improve supervised and unsupervised ML models for fraud detectionHandle imbalanced, noisy real-world data and assess models using fraud-relevant metricsAnalyze false positive / false negative trade-offs and provide model explainability (e.g. SHAP)Collaborate with developers and database teams to integrate ML outputs into the applicationCommunicate results clearly to technical and non-technical stakeholders, including the customerYour Profile:Strong Python for data analysis and machine learning (e.g. pandas, scikit-learn)Advanced SQL for analytical queries and feature generationStrong analytical and structured working styleExperience with supervised learning as well as unsupervised methods such as clustering, outlier detection, or semi-supervised approachesGood understanding of overfitting, regularization, feature or label leakage, and its mitigationHands-on exposure to validation strategies under production conditionsExperience with cloud-based ML software stackCommunication skills to understand business needs and generate business valueNice to Have-Experience with AWS Stack-Familiarity with fraud-specific metrics or cost-sensitive modeling-Experience with Generative AI (e.g. for experimentation, exploration, or documentation)-Basic understanding of software engineering concepts and APIsHow about?Employment contract?Of course. With us you do not have to worry about stable employment.
Benefits?We have them Among other: corporate incentive program, sport card, private medical care.
Lunch card?With the cooperation extended and permanent contract, you will receive additional funds to use for meal purchases.
Working in a hybrid model?Of course You work with us 2 days a week from home.
Work wherever you want?In PAYBACK you have the opportunity. Working 100% remotely, also from European countries for 15 days a year.Flexible
working hours
?Sounds great We start working between 7 to 10.
Trainings?Of course. We provide training to develop hard and soft skills.
Convenient location?Sure We invite you to our new office at Rondo Daszyńskiego, but we are currently also working remotely.
Dress code?We definitely say no. There are no rigid dress code rules in our company, sneakers are more than welcome.
Friendly atmosphere at work?Yes In PAYBACK, people are the most important asset.Something is missing?Open communication is our priority, so dare to ask
The role owns the data science and machine learning aspects of the solution, from data analysis to model evaluation.
The role is well suited for colleagues with strong analytical, SQL, Python, or software backgrounds who want to grow into a machine‑learning‑focused position.
Your
responsibilities:
Analyze transactional and behavioral data to identify fraud patterns; perform data quality checks and exploratory analysisPrepare and document datasets for training, validation, and testing using SQL and PythonDesign fraud-relevant features and translate business logic into measurable signalsDevelop, evaluate, and improve supervised and unsupervised ML models for fraud detectionHandle imbalanced, noisy real-world data and assess models using fraud-relevant metricsAnalyze false positive / false negative trade-offs and provide model explainability (e.g. SHAP)Collaborate with developers and database teams to integrate ML outputs into the applicationCommunicate results clearly to technical and non-technical stakeholders, including the customerYour Profile:Strong Python for data analysis and machine learning (e.g. pandas, scikit-learn)Advanced SQL for analytical queries and feature generationStrong analytical and structured working styleExperience with supervised learning as well as unsupervised methods such as clustering, outlier detection, or semi-supervised approachesGood understanding of overfitting, regularization, feature or label leakage, and its mitigationHands-on exposure to validation strategies under production conditionsExperience with cloud-based ML software stackCommunication skills to understand business needs and generate business valueNice to Have-Experience with AWS Stack-Familiarity with fraud-specific metrics or cost-sensitive modeling-Experience with Generative AI (e.g. for experimentation, exploration, or documentation)-Basic understanding of software engineering concepts and APIsHow about?Employment contract?Of course. With us you do not have to worry about stable employment.
Benefits?We have them Among other: corporate incentive program, sport card, private medical care.
Lunch card?With the cooperation extended and permanent contract, you will receive additional funds to use for meal purchases.
Working in a hybrid model?Of course You work with us 2 days a week from home.
Work wherever you want?In PAYBACK you have the opportunity. Working 100% remotely, also from European countries for 15 days a year.Flexible
working hours
?Sounds great We start working between 7 to 10.
Trainings?Of course. We provide training to develop hard and soft skills.
Convenient location?Sure We invite you to our new office at Rondo Daszyńskiego, but we are currently also working remotely.
Dress code?We definitely say no. There are no rigid dress code rules in our company, sneakers are more than welcome.
Friendly atmosphere at work?Yes In PAYBACK, people are the most important asset.Something is missing?Open communication is our priority, so dare to ask