About the position We are looking for a **Senior Data Scientist** with strong Machine Learning expertise to develop and deploy models that directly impact the business — from demand forecasting to NLP and embedding solutions built on our product catalog and commercial data. You will work closely with business units and the Data Engineering team, transforming real-world problems into reliable, measurable, and production-ready models. This is not just a notebook-based role: we expect you to own the full lifecycle from problem definition to delivery, with careful attention to validation, metrics, and maintenance. **Responsibilities** Develop, validate, and deploy supervised and unsupervised Machine Learning models (forecasting, classification, propensity, segmentation). Build NLP and embedding solutions — for example, product catalog matching and enrichment via semantic similarity, and classification and extraction from textual descriptions. Perform feature engineering and large-scale data preparation directly in the data warehouse. Define appropriate evaluation metrics for each problem, conduct rigorous experimentation (baseline, validation, testing), and clearly communicate results to both technical and business stakeholders. Monitor model performance in production and propose improvements (e.g., concept drift, retraining, optimization). Collaborate with Data Engineering to build reliable pipelines and deploy solutions. Document model assumptions, decisions, and limitations. **Requirements** Proven experience as a Data Scientist with real, production-deployed deliverables. Strong Python skills for data science (pandas, NumPy, scikit-learn, and similar libraries). Advanced SQL proficiency and comfort working with large-scale datasets. Practical expertise in ML modeling: algorithm selection, validation, metrics, and model diagnostics. Experience with NLP and/or embeddings (text vectorization, semantic similarity, language models). Proficiency in version control (Git) and code organization best practices. Critical thinking about data: ability to question assumptions, identify biases, and validate results before relying on them. Strong communication skills — explain trade-offs and results to both technical and non-technical audiences. **Nice-to-have** Experience with Snowflake, especially Cortex (vector functions, LLM functions, Cortex Search). Hands-on experience with multilingual embedding models (e.g., multilingual-e5, bge-m3, BERTimbau) and hybrid approaches (BM25 + dense). Familiarity with the Azure ecosystem and Power BI / Microsoft Fabric. Basic understanding of MLOps (model versioning, monitoring, CI/CD for ML). Experience in demand forecasting and time series analysis. Domain knowledge in consumer goods (FMCG/CPG) and commercial topics such as Revenue Growth Management, pricing, and promotions.
João Silva
Indeed · HR
João Silva
Indeed · HR
