Job Summary:
Client is seeking a hands-on Applied ML / Application Engineer to support a global manufacturing organization focused on process chemical manufacturing. This is a critical role that sits at the intersection of data science, machine learning, and manufacturing operations. The ideal candidate understands how to apply data science principles in a production environment and can work closely with both business and operational teams to drive value through anomaly detection, yield optimization, and process improvements.
Note: Preference is local Curitiba Brazil for sporadic onsite visits.
Responsibilities:
* Collaborate with business stakeholders, manufacturing engineers, and data teams to translate operational challenges into ML-driven solutions.
* Utilize existing ML models (and adjust where necessary) to support yield improvement, root-cause analysis, and anomaly detection.
* Understand contextual asset and process data to build digital threads across manufacturing workflows.
* Act as a bridge between the manufacturing floor and data/ML teams—communicating technical concepts in a way that's digestible to non-technical stakeholders.
* Work with data from multiple systems including MES, QMS, LIMS, process historians, and time series platforms.
* Learn and leverage TwinThread platform to operationalize and deploy ML models.
* Collaborate with engineering teams on data ingestion and pipeline development when necessary.
* Support visualization efforts that convey model insights and drive data-driven decision making.
* Ensure models align with existing ISA-95 / ISA-88 architectures across batch and continuous processing environments.
Experience:
* 3–7 years’ experience in applied ML / data science / data engineering roles, preferably in a process manufacturing or chemical manufacturing setting.
* Background in chemical engineering, industrial automation, or manufacturing systems.
* Knowledge of ISA-95 / ISA-88 standards
* Strong understanding of manufacturing operations, particularly continuous and batch processing systems.
* Proven experience with anomaly detection, predictive maintenance, or yield optimization in an industrial context.
* Familiarity with time-series data, process historians, QMS, LIMS, MES, and related systems.
* Comfortable interacting with both business stakeholders and hands-on operational staff;
strong communication and stakeholder engagement skills.
* Experience working with TwinThread or similar industrial AI platforms a strong plus (willingness to learn is acceptable).
* Experience with data visualization tools and storytelling with data.
* Familiarity with cloud platforms and common ML libraries (e.G., scikit-learn, PyTorch, TensorFlow) is beneficial but not essential.
Nice to Have:
* Experience with model deployment in production systems.
* Exposure to digital twin or digital thread concepts.
* English and Portuguese speaking (they have translators if not)