Data Scientist

Salt; Abu Dhabi; On-site; AED 30,000 - 32,000; Full time

Job summary

Salt is seeking a hands-on Senior Data Scientist with banking or financial services experience, particularly in credit risk and lending. You will develop machine learning solutions for loan underwriting, default prediction, and customer analytics, with ownership across model development, validation, deployment and monitoring.

The ideal candidate should have strong technical fundamentals and be able to explain modelling decisions and business outcomes clearly.

Qualifications

• Proven ML project delivery with ownership of decisions and outcomes.
• Banking/financial services experience with credit risk or lending modelling.
• Strong Python/SQL skills for data prep, analysis and modelling.

Responsibilities

• Translate business needs into modelling objectives with measurable success criteria.
• Develop models for credit scoring, underwriting, and default prediction.
• Analyze credit bureau, transactional and behavior data to improve risk assessment.
• Perform data prep, feature engineering, and feature selection.
• Build and compare models using Random Forest, XGBoost, LightGBM, CatBoost.
• Conduct hyperparameter tuning and validation such as cross-validation and time-based tests.
• Evaluate models and select thresholds aligned with business priorities.
• Explain model behavior and individual predictions using SHAP.
• Collaborate with engineering to deploy models and monitor performance.

Skills

Python

SQL

Credit risk modelling

Machine learning

Model deployment

Data analysis

Tools

AWS SageMaker

XGBoost

LightGBM

CatBoost

SHAP

Job description

Salary – AED 30,000 – AED 31,500 per month

Contract Length – 6 Months (view to extend)

Start Date – ASAP

The customer is specifically looking for a Data Scientist (Not AI Engineer) with banking sector experience and strong hands-on exposure to credit risk-related projects, ideally with around 6 to 8 years of experience. This includes work on credit risk, loans, scoring platforms, scorecards, underwriting, or similar use cases. Strong data science depth is essential.

Role Overview

We are looking for a hands-on Senior Data Scientist with banking or financial services experience, particularly in credit risk and lending. The role involves developing machine learning solutions for loan underwriting, default prediction, and customer analytics, with ownership across model development, validation, deployment, and monitoring.

The ideal candidate should have strong technical fundamentals and be able to explain modelling decisions and business outcomes clearly.

Key Responsibilities

• Translate business requirements into modelling objectives, precise target definitions, and measurable success criteria.
• Develop models for credit scoring, loan underwriting, probability of default and borrower behavior analysis.
• Analyze credit bureau, transactional, repayment and behavioral data to improve risk assessment.
• Perform data preparation, feature engineering, and feature selection.
• Build and compare classification models using algorithms such as Random Forest, XGBoost, LightGBM and CatBoost.
• Conduct hyperparameter tuning and apply appropriate validation methods, including cross validation and out of time testing.
• Evaluate model performance and select classification thresholds aligned with business priorities.
• Explain overall model behavior and individual customer predictions using techniques such as SHAP.
• Collaborate with engineering teams to deploy models and monitor performance, stability and data drift.
• Present model results, limitations and measurable business impact to technical and business stakeholders.

Required Skills and Experience

• Proven experience delivering machine learning projects, with clear ownership of technical decisions and outcomes.
• Banking or financial services experience, including hands on modelling for credit risk or lending.
• Understanding of loan products, underwriting, default definitions, repayment behavior, and credit bureau data.
• Strong proficiency in Python and SQL for data preparation, analysis, and modelling.
• Strong understanding of classification algorithms, bagging, boosting and the bias variance trade off.
• Practical knowledge of ROC AUC, KS, precision, recall, F1, confusion matrices and classification threshold selection.
• Experience with feature selection, correlation analysis, normalization, standardization, and model explainability.
• Strong understanding of data leakage, overfitting, underfitting, dataset splitting, and final model training.
• Ability to communicate technical concepts clearly and connect model performance to business KPIs.

Preferred Skills and Experience

• Experience with AWS and Amazon SageMaker for model training, deployment, and monitoring.
• Experience with hyperparameter optimization tools such as Optuna.
• Experience developing customer propensity, recommendation, or cross selling models for financial products.
• Experience combining traditional credit data with alternative or behavioral data.
• Banking experience and hands-on AWS SageMaker experience