6h ago

Cientista de Dados Especialista (Modelagem de Crédito)

BrazilRemoteFull-timemid

OtherCientista De Dados

1 views0 saves0 applied

Apply Now

Quick Summary

Overview

This position is listed on behalf of a partner company, who manages all applications and next steps.

Technical Tools

OtherCientista De Dados

As a Senior Data Scientist, you will serve as a technical reference for credit modeling, leading the full lifecycle of statistical and machine learning models.
You will work with large volumes of structured and unstructured data to develop highly predictive solutions across the customer credit lifecycle.
The role combines advanced modeling, data engineering, validation, deployment, and continuous production monitoring.
You will help optimize the balance between credit risk and return by turning complex data into actionable insights for decision-makers.
You will collaborate closely with teams across Credit, Policy, Risk, Products, Fraud Prevention, and Engineering.
This is a strategic and hands-on opportunity to introduce new modeling techniques, alternative data sources, and AI capabilities into credit decision-making.
You will also contribute to technical standards, model governance, reproducibility, and the continuous evolution of the data science function.

Develop and maintain the technical playbook and model development pipelines, standardizing methodologies, reusable functions, validation criteria, and structured outputs for executive and technical presentations while ensuring efficiency, reproducibility, and compliance with technical requirements.
Lead the creation, calibration, and evolution of predictive models across the credit lifecycle, including Application, Behavior, Collection, Early Warning, PD, LGD, and EAD models.
Own the end-to-end modeling process, from data extraction and manipulation and feature engineering through methodology definition, validation, deployment support, and implementation in collaboration with Engineering and IT teams.
Continuously evaluate model performance and stability in production using metrics such as Gini, KS, and Population Stability Index (PSI), recommending recalibration or replacement when appropriate.
Create and maintain technical model documentation and model dossiers, ensuring auditability, reproducibility, and alignment with business and regulatory requirements.
Explore, test, and implement new statistical techniques, Machine Learning algorithms, and alternative data sources to improve model discriminatory power.
Act as a strategic analytical partner to Policy, Risk, and Fraud Prevention teams, providing risk curves, simulations, and analytical insights to support executive decision-making.
Identify opportunities to expand the use of Artificial Intelligence across the team's workflows, from project ideation and management through technical execution of analytical initiatives.

Requirements

~1 min read

Proven and consistent experience building and validating credit risk models within financial institutions, fintechs, credit bureaus, or specialized consulting firms.
Advanced proficiency in programming languages used for data analysis and statistical modeling.
Strong experience with languages and technologies for extracting and manipulating large volumes of Big Data, including structured and unstructured datasets.
Solid theoretical knowledge and practical experience with advanced Machine Learning and Deep Learning techniques, logistic regression, decision trees, validation, and sampling methodologies.
Familiarity with model explainability frameworks for black-box models, including SHAP and LIME.
Deep understanding of the credit lifecycle, fundamental risk metrics, and the impact of analytical solutions throughout the credit cycle.
Strong strategic perspective, with the ability to identify, structure, and expand the use of Artificial Intelligence across data science workflows.
Ability to translate complex analytical findings into practical inputs for business and executive decision-making.
Strong problem-solving, analytical, and technical communication skills, with the ability to work collaboratively across multidisciplinary teams.