Description

The Role: 
Xylem’s Decision Intelligence business unit is seeking candidates with a background in Computer Science or Data Science and relevant practical experience developing software products to join our team as an MLOps Engineer in North America. As a Data Science team member, the incumbent will design, develop, and test software products for water systems to integrate with data services and support Digital Intelligence solutions and technologies. The successful candidate will demonstrate the ability to develop scalable, performant software products that provide clear value to the water industry. The ideal candidate will clearly understand the data science product lifecycle and delivery of production services. Candidates who also have valuable experience in the water analytics space are especially encouraged to apply, including water infrastructure systems. The Xylem data science team is deeply involved in developing online predictive models and digital twins of water/wastewater treatment processes and water distribution and collection systems, using various numerical analysis and optimization algorithms and AI/machine learning models; the successful candidate will demonstrate a clear ability to contribute to and lead, these ongoing and related new product development efforts.

Essential Duties/Principal Responsibilities:

• Develop and maintain automated machine learning pipelines, streamlining the model lifecycle from training and validation to deployment.
• Configure data science products for deployment within the AWS or Azure ecosystem, and set up MLOps technologies for sub-routine tasks, such as model deployment and retraining.
• Utilize containerization (e.g., Docker) and orchestration (e.g., Kubernetes) to ensure efficient model deployment and management.
• Apply software engineering best practices, including CI/CD and automation, to machine learning.
• Contribute to the development of REST APIs in Django for efficient model integration into production applications.
• Create technical documentation to support software product development using formats like OpenAPI, Markdown, and LaTex.
• Implement best practices for data science, including code development, model training, and validation/testing using MLOps.
• Collaborate closely with data scientists to ensure reliable, scalable, secure, and consistent model deployment.
• Build unit tests to ensure the accuracy and robustness of all model components.
• Demonstrate a strong sense of ownership to complete all product deliverables.

Minimum Qualifications: Education, Experience, Skills, Abilities, License/Certification:

• Bachelor’s degree in computer science or software engineering
• A background of at least 2 years in relevant applied work experience
• Strong programming skills in languages such as Python
• Familiar with relevant machine learning and scientific libraries (e.g., TensorFlow, Pytorch, scikit-learn, pandas)
• Strong understanding of Docker and container orchestration systems like Kubernetes or ECS
• Understands the ins and outs of relational databases, including schema design
• Ability to design and implement cloud solutions (AWS, MS Azure or GCP)
• Strong understanding of Rest APIs with experience in a framework like Django, Flask, or FastAPI
• Understanding of software development principles and practices, including version control (e.g., Git) and software testing
• Proficient with CI/CD, including tools like Bitbucket Pipelines or Github Actions
• Possesses advanced critical thinking skills, enabling objective evaluation and selection of diverse tools, platforms, and technologies to construct an efficient and dependable MLOps tech stack
• Ability to harness collective expertise and foster a cooperative spirit among team members
• Has an always growing mindset
• Fluent in English, with excellent communication skills

Preferred Qualifications:

• Proven experience in the end-to-end lifecycle management of machine learning models, encompassing model development, testing, deployment, monitoring, and continuous improvement
• Experience with MLOps tools and frameworks like Kubeflow, MLFlow, Sagemaker, Airflow etc.
• Experience with Django and its ecosystem
• Proficiency in IaC Tools, such as Terraform, CloudFormation, Ansible, or similar