Sr Data Scientist (LATAM Remote)

Technical Challenge:

In This Role, You Will

• Apply statistical and machine learning methods to operationally meaningful problems.
• Build and refine digital twins and predictive models of physical assets, processes, and operational workflows.
• Work across hybrid data estates that span on-prem operational systems and modern cloud platforms.
• Use modern ML frameworks (PyTorch, TensorFlow) where they earn their place, and simpler tools where they don't.
• Run rigorous, reproducible experimentation using tools like MLflow.
• Work with large-scale structured and unstructured datasets in Snowflake environments
• Develop and maintain scalable data pipelines, ETL/ELT workflows, and ML infrastructure
• Design systems for storing, processing, and managing ML outputs, embeddings, and AI-generated data
• Collaborate with cross-functional teams to translate business problems into scalable data solutions

Required Skills & Expertise

• Hands-on experience in Data Science, Machine Learning, or ML Engineering roles in Big Data environments
• Strong applied statistics: experimental design, inference, uncertainty quantification, and a working sense of when a result is real versus an artifact of the data.
• Strong Python and SQL fluency, including comfort with modern distributed SQL engines (e.g., Trino, Spark SQL, or similar).
• Comfort working across hybrid data environments spanning on-prem operational sources and modern cloud platforms (AWS, Azure, or GCP).
• Experience with the full ML lifecycle: ingestion, transformation, feature engineering, training, evaluation, deployment, and monitoring.
• Practical experience with modern ML frameworks (PyTorch or TensorFlow) and experiment tracking tooling (MLflow or comparable).
• Strong problem-solving skills and ability to work in fast-paced startup environments
• Experience working with Snowflake in production data environments.

Nice to Have

• Experience delivering models as containerized services like Docker and Kubernetes
• Direct experience with digital twins or applied modeling of physical / operational systems.
• Time-series, sensor, or streaming data at production scale.
• Open lakehouse formats (Iceberg, Delta, Hudi) and table-format-aware workflows.
• Causal inference, A/B testing, or sequential evaluation
• Edge or hybrid model deployment patterns.
• Experience with Databricks or comparable platforms.

UP.Labs Summary