Senior Data/ML Engineer — Databricks Forecasting Platform | Remote

Das ist der Job

Louis, but open to remote consultants globally with North American working-hours overlap

Darum lohnt es sich

This is not a greenfield build — the focus is modernizing a Databricks-based forecasting system that's been maintained primarily by a single developer for years.

Headquarters: Remote URL: https://www.toptal.com/ About the Role We're looking for a Senior Data/ML Engineer to refactor, operationalize, and improve an existing time series forecasting platform that's already live and driving real business value.

You'll reduce technical debt, strengthen testing and observability, and raise the engineering bar on a system the business already depends on. If you'd rather bring discipline and maturity to an existing production system than start from a blank slate, this is built for that.

What You'll Do • Review the current forecasting platform architecture and identify areas for improvement • Refactor existing Databricks, Python, and PySpark implementations • Move business logic out of Databricks notebooks and into reusable Python modules or packages • Improve separation of concerns between orchestration and core business logic • Establish stronger engineering standards and help define what "good" looks like for the platform • Implement or improve automated testing practices and validation mechanisms for forecasting workflows • Build or improve monitoring and observability, increasing visibility into how predictions are generated • Help monitor model behavior and operational health over time • Improve reliability of scheduled training workflows, reducing manual intervention on failure • Improve failure handling, retries, and overall workflow resilience • Maintain and extend existing forecasting capabilities as needed What You Bring • Strong professional experience with Databricks, including workspaces, notebooks, scheduled workflows, and CI/CD processes • Strong Python engineering experience, including designing reusable modules or packages • Strong PySpark experience with production data pipelines or distributed data processing • Experience refactoring production code and improving maintainability • Familiarity with time series forecasting concepts and workflows • Ability to understand and work effectively within an existing, unfamiliar codebase • Experience improving software quality, testing strategy, and engineering standards • Experience implementing automated testing practices • Experience improving monitoring, observability, or operational visibility for production systems • Strong judgment around technical debt, refactoring priorities, and maintainable architecture • Ability to work with existing systems rather than only building from scratch Why This Role • Real production impact: Improve a system the business already relies on, not a proof-of-concept • Engineering maturity focus: Bring testing, observability, and maintainability to a platform that’s outgrown its current state • Meaningful ownership: Help define engineering standards for the forecasting platform going forward • Flexible location: Preference for Toronto or St.