Data Engineer – Industrial data contextualization

Project

Build the information layer that explains what each industrial signal means and connects it to the correct machine, component, operating context, and maintenance record. The goal is data Nexa can interpret consistently across different machines and sites.

Responsibilities

• Define a practical data model for assets, components, signals, units, sources, timestamps, operating states, alarms, and data-quality status.
• Build or improve Python and SQL transformations that normalize incoming telemetry and preserve its original source and time.
• Map signals to asset-register information and, where available, alarms, production events, work orders, preventive tasks, and parts records.
• Design validation checks for missing or stale readings, duplicates, timestamp problems, counter resets, unit mismatches, and implausible values.
• Create a small contextualization prototype and document a repeatable onboarding method for a newly connected machine.

Expected deliverables

• Industrial data model and signal dictionary: Definitions for asset/component, signal meaning, unit, source, sampling, expected range, and quality status.
• Working transformation pipeline or prototype: Reproducible ingestion and normalization using an agreed test dataset or authorized connection.
• Contextualized demonstration: At least one agreed machine or test asset with mapped signals and links to relevant operational records where available.
• Data-quality report and onboarding guide: Documented validation results, known gaps, and steps to add a new machine or signal.

Profile and skills

Final-year engineering or equivalent master’s student in data engineering, computer science, software engineering, or a related field. Comfort with Python, SQL, data modeling, and data transformation, with an interest in time-series data and industrial operations.

Useful experience

PostgreSQL, pandas, REST APIs, AWS, Docker, OPC UA, MQTT, Modbus, and maintenance datasets.

Shared program context

This is one of three coordinated end-of-studies projects contributing to a secure and context-aware industrial data integration path. The shared outcome is to connect industrial data to the right machine and operating context and make it useful to Nexa through a secure, repeatable integration path.