Machine Learning Engineer

We are the people who give possibilities purpose BD is one of the largest global medical technology companies in the world. Advancing the world of health™ is our Purpose, and it’s no small feat. It takes the imagination and passion of all of us—from design and engineering to the manufacturing and marketing of our billions of Med Tech products per year—to look at the impossible and find transformative solutions that turn dreams into possibilities.

As a Staff Machine Learning Engineer, you will lead the design, development, and deployment of machine learning solutions that power AI-driven decision-making across the BD Connected Care portfolio. You will build and operationalize models for forecasting, optimization, anomaly detection, and other advanced analytics use cases, transforming high-quality data into scalable, intelligent products. This is a hands‑on technical leadership role where you will own the end‑to‑end machine learning lifecycle, from experimentation and model development to deployment, monitoring, and continuous improvement.

You will help define the architecture, engineering standards, and best practices that shape the future of BD's AI platform while collaborating across engineering, product, and clinical teams to deliver impactful solutions. The ideal candidate combines deep machine learning expertise with strong software engineering skills, takes ownership of complex technical challenges, and is motivated by BD's mission to improve patient outcomes through innovation and AI.

• Modeling Ownership:
  Own the design, training, and validation of demand forecasting, optimization, and anomaly detection models across product, operational, and clinical‑adjacent data. Build probabilistic, multi‑series models that emit prediction intervals rather than point estimates, so recommendations can be set to a defined service‑level target.
• Methods Breadth:
  Apply classical methods (ARIMA/ETS/state‑space, Bayesian, gradient boosting, hierarchical forecasting, LP/MIP/convex optimization) and deep learning approaches (transformer‑based and other neural forecasting architectures, foundation‑model adaptation), choosing based on the problem.
• Training and Experimentation:
  Own training pipelines end to end, including feature engineering, dataset versioning, and reproducibility. Build rigorous experimentation with temporally aware splits, honest baselines against the incumbent rule‑based system, ablations, calibration, and uncertainty quantification. Design back tests and shadow‑mode evaluations that precede any production trust.
• Production:
  Turn models into production services on AWS (EKS, S3, Lambda, Event Bridge, Step Functions): inference services, batch jobs, and data integrations. Implement clean, modular, well‑tested code with type hints, unit and integration tests, dependency hygiene, and reproducible builds.
• Evaluation and Monitoring:
  Implement model evaluation and continuous‑evaluation harnesses, instrument production inference for quality, latency, and drift, and partner with ML Platform on packaging, deployment, promotion gates, and rollback.
• Explainability:
  Develop explainable outputs that surface why a recommendation was made, so users can accept, override, or investigate with confidence.
• Mentorship and Standards:
  Mentor AI and ML engineers, review designs and code, and contribute to engineering standards and team‑wide tooling.
• Documentation:
  Document methodology, assumptions, and validation in Confluence; produce design docs, ADRs, runbooks, and model cards suitable for regulatory and clinical review. Documentation is a deliverable, not an afterthought.

• Bachelor's in Machine Learning, Statistics, Operations Research,…