Machine Learning Engineer

Senior ML / Data Engineer | Sydney, AustraliaThis is not a research role. It is a hands-on engineering role for someone who writes production code, owns the model lifecycle end to end, and takes satisfaction in seeing something they built running reliably in the field generating real commercial outcomes.

What You Will OwnTaking the existing predictive maintenance model to production. Your job is to improve the accuracy, harden the training data strategy, validate the model rigorously, deploy it properly on the existing AWS platform, and set up monitoring, drift detection, and retraining cadence so it continues to perform in the field.

Building and maintaining ML-ready data pipelines. The platform runs on AWS with S3, Glue, Airflow, Step Functions, Snowflake, and dbt. You will own the ML feature engineering layer on top of that infrastructure, including training and serving pipelines, data quality validation, and production monitoring feeds.

Working with time-series IoT device telemetry data. The connected devices generate continuous operational event streams from clinical sites globally. You will transform raw telemetry into features a model can learn from, handling real-world data challenges including missing signals, irregular intervals, rare failure events, and imbalanced training sets.

What We Are Looking For

You have personally trained, validated, and deployed at least one model in a live environment with a quantifiable business outcome. You can speak to the training data strategy, model selection, evaluation methodology, accuracy trade-offs, and how the model performed over time in production.

Time-series or sensor data experience. Experience with IoT device telemetry, predictive maintenance, anomaly detection, or sequential operational data is a strong advantage. The core problem here is supervised classification and regression on continuous device event streams.

Strong AWS data and ML stack. Hands-on with SageMaker, S3, Glue, Airflow, and Step Functions.MLOps fundamentals. You understand model drift, retraining triggers, production monitoring, and the operational discipline required to keep a model performing reliably after deployment.

Experience with D365 ERP data or other enterprise source systems is a genuine advantage given the data environment this role operates in.

If interested, please apply with your most up-to-date CV and I will reach out.