Editar oferta

Título del puesto *Empresa *Portal de origenURL de la ofertaUbicaciónPaís (código ISO)Fecha de publicaciónModalidad

Salario mín.Salario máx.Moneda

Descripción

Hi, we're Brigit! A holistic financial health company helping everyday Americans build a brighter financial future. With a business model that is aligned with our customers, we create transparent, fair, and simple financial products that put money back in the hands of our members, help them spend wisely, avoid unfair fees and build their credit quickly. If autonomy, ownership, and having meaningful input at the company you work for is important to you, come join our growing team! Brigit is doing innovative and exciting work, but donât just take our word for it, our work is being recognized by others: Built In's 2026 Best Midsize Companies to Work For in New York City Built In's 2025 Best Places to Work For In New York City Built Inâs 2024 & 2025 Best Startups to Work For In the U.S. Built Inâs 2023 - 2025 Best Startups to Work For In New York City Fast Companyâs Most Innovative Companies of 2022 Business Insiderâs Most Promising Consumer Startups 2022 Forbes Fintech 50 2022 Role Overview: At Brigit weâre focused on giving the 100M Americans who live paycheck to paycheck access to more affordable financial services and getting them on the path to better financial wellness. As a Data Scientist on our team youâll be responsible for building, improving and maintaining the key ML models that enable our services. The key problem spaces we are focused on are: Identifying credit risk of customers so that we can help more people in need when they need it Identify fraud early, so we can focus on customers that rely on us for support Optimizing our transaction timing so we arenât causing our customers to overdraw their accounts. We have access to rich, structured data that we can use to derive insights and build complex models. In addition to supporting new use cases when needed, you will also have the opportunity to help us stand-up best practices and collaborate across Data Science. Analytics, Product and Engineering teams. What youâll be doing: Build, test and roll out new underwriting and risk models to better predict risk Youâll have ownership of the full modeling lifecycle from conception through to production and realize every bit of impact along the way. Build, test and roll out new or improved models related to fraud and payments. Contribute to building best practices for feature development, model training and model testing and monitoring. Analyze how our customer base is shifting as we grow and pinpoint areas we can improve. Help our existing engineering and business teams achieve our cross-functional goals. Mentor more junior data scientists or aspiring data scientists across the data team. What you have: 6+ years of experience doing data science (modeling + analysis) Advanced degree in data science, statistics, computer science, or related field. Proven expertise in machine learning principles and techniques Experience in Python, using industry-standard modeling toolkits like sklearn, JupyterLab, pandas, matplotlib, statsmodels, etc. Experience in writing complex SQL queries and the ability to put together multiple data sources Experience creating training data sets, training models, tuning hyperparameters, performing validation, running A/B tests in production, creating and owning performance and model monitoring Experience building classification and prediction models, testing them in a startup environment and iterating to improve their models. Ability to work effectively and communicate ideas/code clearly in a cross functional team environment (Engineering, Product and Business teams) Experience building end-to-end data science solutions to address business needs: from ideation to deployment & AB testing Strong written and verbal communication skills, with the ability to explain complex concepts to various audiences Who You Are Passionate self-starter who is always looking for ways to optimize, test, and improve Ability to operate through ambiguity by asking questions and learning quickly Always willin

Requisitos (uno por línea)Keywords (separadas por comas)