Python, MLOps

Role Summary We are seeking a Software Engineer with strong Python and MLOps foundations to build enhance and stabilize a production-grade AI/ML platform supporting model training validation deployment and orchestration at enterprise scale. This role goes beyond scripting or notebook-driven ML. You will work in a class-based API-driven CI/CD-enforced MLOps ecosystem contributing to reusable libraries ML workflows secure endpoints JSON-schema-driven interfaces and automated pipelines aligned with Chevrons evolving GitHub Actions strategy. Key Responsibilities Core Engineering & MLOps - Design build and maintain production-grade Python services using sound object-oriented principles (SOLID separation of concerns reusability). Contribute to and extend the enterprise MLOps pipeline supporting: - Model training validation and registration - Azure ML based execution - Parameterized workflow orchestration - CI/CD-driven deployments Implement and maintain ML inference endpoints with focus on: -High-performance I/O using async programming and concurrency -Clean request/response contracts driven by JSON schemas -Robust validation and error handling Develop well-defined APIs (REST) with: -OpenAPI / Swagger documentation -Version-aware schemas and backward compatibility considerations Support platform evolution from Azure Pipelines to GitHub Actions contributing to: -Pipeline re-architecture -Build test and release automation -Secure artifact promotion across environments Data ML & Storage Work with Pandas and Polars for feature handling transformations and data preparation. Support ML workflows leveraging Scikit-Learn models and pipelines.

Integrate with Azure Blob Storage and Azure Data Lake for model artifacts datasets and metadata. (Nice to have) Contribute to solutions involving Azure Cosmos DB for metadata or workflow state tracking. Quality Testing & Maintainability Write clean maintainable and testable code-optimizing for long-term platform health over short-term delivery speed. Expand and strengthen the automated test suite including: -PyTest-based unit and integration tests -Validation of pipelines schemas and services Advocate for and gradually adopt Test-Driven Development (TDD) practices across the codebase. Actively reduce technical debt in a legacy-heavy environment by: -Refactoring duplicated or brittle code -Introducing shared libraries and abstractions -Improving documentation and developer ergonomics Required Technical Skills Advanced Python -Class-based design modular architecture reusable components -Not script-only or notebook-centric development API Development -RESTful services endpoint performance tuning -OpenAPI / Swagger documentation Async Programming -Async/await concurrency threading for I/O-heavy workloads CI/CD -Azure Pipelines and/or GitHub Actions -Experience evolving pipelines not just consuming them Data & ML -Pandas / Polars -Scikit-Learn Testing -PyTest -Strong appreciation for automated testing as a quality gate Schema-Driven Development -JSON schemas for configuration workflow parameters and APIs

-Experience updating and managing schema evolution Nice-to-Have Skills Azure ML SDK (training pipelines model registration) Pydantic for request/response validation Azure Cosmos DB Schema versioning strategies Experience with large shared enterprise codebases Desired Qualities & Mindset Takes Initiative -Understands how individual stories connect to platform-level objectives. -Proactively improves systems instead of waiting for direction. Strong Debugger -Comfortable tracing failures across pipelines services storage and infrastructure. -Attempts root-cause analysis before escalating issues. Values Code Quality -Will not compromise maintainability for speed. -Actively resists adding to technical debt. Engineering Craftsmanship -Cares about design clarity and long-term scalability. -Sees MLOps as software engineering not just ML enablement. Advocates for Better Practices -Encourages testability consistency and clean abstractions-even in legacy environments.