Senior Data Scientist

Description

Job Summary: Join the Analytics & Data Science team to develop statistical and machine learning models that optimize commercial decisions and portfolio growth. Key Highlights: 1. Develop advanced statistical and machine learning models 2. Focus on sales optimization, customer retention, and offer personalization 3. Work with MLOps, Databricks, Spark, and AWS **Your Responsibilities:** * Join the Analytics \& Data Science team to develop statistical and machine learning models supporting commercial decisions, optimizing sales channels, increasing conversion and retention, and growing the portfolio — with scientific rigor, controlled experimentation, and large-scale production models. **Responsibilities and Qualifications** **What You Should Have:** * Build **Next Best Action** models to recommend the **ideal product, channel, and timing**, increasing **cross\-sell** across the customer base and **conversion** on digital channels. * Develop **churn** models using **behavioral and policy history**, anticipating **cancellation risk** and triggering **personalized retention actions**. * Deliver **purchase propensity** models for **lead scoring and active customer base**, prioritizing customers by **likelihood of contracting**, guiding **commercial and digital outreach**. * Generate **behavioral and economic segmentation** to **personalize communication, offers, and pricing** by profile. * Apply **uplift modeling** to measure the **incremental impact of campaigns**, focusing where there is **real behavioral change**, reducing **budget waste**. * Design and analyze **A/B tests and RCTs** to measure the **causal impact** of sales initiatives, campaigns, and journey changes — ensuring **evidence-based decisions**. * Build **Marketing Mix Modeling (MMM)** and **multichannel attribution** to measure **channel ROI** and optimize **budget allocation** based on **incrementality**. * Evaluate policies without RCTs using **causal methods**, estimating the **impact of changes**, such as regional pricing adjustments. * Model **predictive LTV** per customer/segment, guiding **acquisition and retention investment**, and supporting the **CAC vs. LTV** logic. * Build and maintain **MLOps pipelines on Databricks**, ensuring **model stability and performance in production**. * Translate analyses into **clear executive recommendations**, connecting **statistical metrics** to **business decisions**. **Requirements and Qualifications** * Bachelor’s degree completed (Statistics, Mathematics, Engineering, Computer Science or related fields) * Advanced proficiency in Python (pandas, numpy, scikit\-learn) and SQL. * Production experience with **Apache Spark** / PySpark / Databricks / SparkML * Experience with **AWS** (S3, Glue, SageMaker, Lambda or similar) in large-scale data environments. * Solid knowledge of **MLflow**: tracking, model registry, champion/challenger aliases, Unity Catalog, nested runs, custom metrics, and artifacts. * Experience with Databricks Feature Store (offline and online tables, Unity Catalog, point\-in\-time correctness, on\-demand features, and streaming features). * Proficiency in MLOps: dev/test/prod lifecycle**, Databricks Asset Bundles (DABs),** model serving endpoints, testing (unit and integration) in notebooks, **automated retraining** with data drift and performance degradation detection. * Batch, real\-time, and streaming inference (including Delta Live Tables) and model deployment to endpoints. * Ability to formulate business problems as well-defined statistical / ML problems. * **Preferred Qualifications** * Prior experience in financial services, insurance, or actuarial domains. * Master’s or Ph.D. in Statistics, Econometrics, Mathematics, Actuarial Sciences, or quantitative fields. * Advanced knowledge of causal inference: DiD, RDD, instrumental variables (IV), propensity score; experience with observational studies and clinical/quasi-experimental trials. * Distributed training and distributed hyperparameter tuning: pandas Function APIs / UDFs, Optuna \+ MLflow, Ray, and understanding of vertical vs. horizontal scaling for ML workloads. * Lakehouse Monitoring: drift detection, statistical tests, time series and snapshot tables, custom metrics, and alerting. * Delta Live Tables and streaming pipelines; custom PyFunc models, REST API, and MLflow Deployments SDK. **Additional Information** **Additional Information** **How We Care For You:** **Benefits:** * Health Insurance; * Dental Insurance; * Life Insurance; * Transportation Allowance; * Meal Voucher; * Food Allowance; * 13th Food Basket; * Group Life Insurance; * Private Pension Plan; * Daycare and/or Nanny Assistance; * Profit Sharing and Results Program (PLR); * Gympass; * Birthday Day Off; * Corporate University with over 500 courses; * Integrated Health Program: mental, social, financial, and physical wellness. **Where Are We Located?** Our headquarters are located in the south zone of São Paulo, near Morumbi train station (Line 9 – Esmeralda), in the Chácara Santo Antônio neighborhood. We are Brasilseg, **a BB Seguros company**, a holding that consolidates the **Bank of Brazil’s insurance businesses.** We operate in Life, Residential, Rural, and Mass-market segments (Residential, Commercial, Condominium, and Machinery & Equipment), being the **national market leader in all segments we serve**. More than 2\.100 employees, distributed across São Paulo (Headquarters), Franca (Customer Relationship and Business Center), and other states (Commercial Team), form a specialized team dedicated to care — working every day to deliver simple, agile, and convenient solutions, **caring for people and protecting what matters most to them**, at every stage of their lives.