Xylem is a Fortune 500 global water solutions company dedicated to advancing sustainable impact and empowering the people who make water work every day. As a leading water technology company with 23,000 employees operating in over 150 countries, Xylem is at the forefront of addressing the world's most critical water challenges. We invite passionate individuals to join our team, dedicated to exceeding customer expectations through innovative and sustainable solutions.
What You’ll Do (Responsibilities)
- Lead end‑to‑end data science projects from scoping through deployment and monitoring—covering telemetry‑based predictive models, LLM agents, and optimization of existing ML solutions.
- Translate business requirements from service operations, product management, and field teams into testable hypotheses, features, and measurable outcomes.
- Develop time‑series models (forecasting, anomaly detection, event prediction) using Python; design features from high‑volume IoT data and join with ERP/CRM context.
- Build and productionize models in Azure: train/evaluate in Azure Machine Learning, register/promote in the AML model registry, and automate with Azure Data Factory/Databricks workflows.
- Engineer robust data connectors (e.g., SQL/pyodbc) to ERP, telemetry stores, and SharePoint/OneDrive sources; use service principals and managed identities for secure refreshes and pipelines.
- Design, evaluate, and ship LLM agents that use retrieval and tool invocation (e.g., RAG over ADLS/Power BI/Dynamics or SAP context) with responsible‑AI guardrails in Azure AI Foundry.
- Instrument model performance & business impact (drift, stability, precision/recall, lead time gain, false‑positive cost) and partner with operations to drive adoption.
- Partner across WSS (DCC operations, IoT software, product, and field service) on SAP‑integrated and Link2Site‑enabled workflows to ensure solutions are operable and auditable.
- Tell the story: build concise Power BI views/dashboards to explain models, KPIs, and value to stakeholders; create clear docs and runbooks.
Required Qualifications
- 3 years experience of applied data science in production environments, preferably in IoT/industrial telemetry.
- Education (Required): Bachelor’s degree in Data Science, Computer Science, Statistics, Applied Mathematics, or a related quantitative field.
- Education Preferred: Master’s degree or higher in a relevant discipline (e.g., Data Science, AI/ML, Industrial Analytics) or equivalent practical experience delivering production-grade ML solutions.
- Strong Python development experience; familiarity with PySpark a plus.
- Time‑series modeling expertise (forecasting, anomaly detection, state models, survival/event prediction; e.g., ARIMA/Prophet, gradient boosting, LSTM/Temporal CNNs).
- Proficiency in SQL and pyodbc; schema design, query optimization, and joining telemetry with ERP/CRM data.
- Hands‑on with Azure ML (training, pipelines, environments), AML model registry, ADLS Gen2, Azure Data Factory, Databricks.
- Practical MLOps: model versioning, CI/CD, orchestration, monitoring, and rollback strategies; Git/GitHub/Azure DevOps.
- Power BI for analytical storytelling and operational monitoring (row‑level security, dataflows, refresh strategies; working knowledge of service principals is a plus).
- Demonstrated ability to translate functional requirements into models and measurement plans; strong stakeholder communication.
Preferred Qualifications
- Education (Preferred): Master’s degree or higher in a relevant discipline (e.g., Data Science, AI/ML, Industrial Analytics) or equivalent practical experience delivering production-grade ML solutions.
- Experience building ML solutions within Databricks or Snowflake.
- Experience with LLM agent design (prompting, tools, RAG, evaluation) in Azure AI Foundry; understanding of guardrails, content filters, and privacy.
- Experience with SAP S/4HANA objects relevant to service (orders, maintenance plans) and integration patterns between IoT platforms (Link2Site) and ERP.
- Domain experience in predictive maintenance, service order automation, or consumables/exchange forecasting within industrial systems.
Example Problem Areas You’ll Tackle
- Predictive models from IoT telemetry to anticipate failures/exchange events and generate proactive service actions; integrate with SAP/field workflows.
- LLM agents that unify knowledge from telemetry, ERP, and SharePoint/OneDrive docs to assist DCC analysts in triage and resolution (RAG + tool use).
- Optimization of existing ML (feature engineering, retraining cadence, rule‑plus‑model ensembles) to improve precision and reduce operational toil.
- Data health & identity: secure refreshes and managed identity patterns for analytics artifacts and pipelines.
Tech Stack You’ll Use
Python, SQL, Azure ML, AML Registry, ADLS Gen2, Azure Data Factory, Azure AI Foundry, Databricks (PySpark), Miccrosoft Power BI/Fabric, Git/Azure DevOps.Join the global Xylem team to be a part of innovative technology solutions transforming water usage, conservation, and re-use. Our products impact public utilities, industrial sectors, residential areas, and commercial buildings, with a commitment to providing smart metering, network technologies, and advanced analytics for water, electric, and gas utilities. Partner with us in creating a world where water challenges are met with ingenuity and dedication; where we recognize the power of inclusion and belonging in driving innovation and allowing us to compete more effectively around the world.
