Automating machine operation in manufacturing requires extensive data analysis. Our technology extracts up to 15 GB of structured manufacturing data per day and per machine. Your task will be to build algorithms that are reproducible for different types of machinery for typical errors and will detect relevant anomalies within the data.
A couple boxes to tick:
- Passionate about Data Science and extracting valuable insights out of complex data structures.
- You have experience in the data science project lifecycle, from data exploration, data cleaning, and feature engineering to building and tuning models.
- You have deployed models to production and maintained them.
- You have a strong background in statistics and an analytical mindset.
- You can explore small datasets quickly but are also comfortable in working with big datasets using batch and streaming data processing jobs.
- You can visualize results of a project in an interactive dashboard for example using bokeh or dash.
- You have a high appreciation for clean architecture and clean code
- You are not afraid of working with challenging datasets and come up with creative methods to overcome obstacles.
- You stay up to date with new developments in machine learning and data science, and are eager to learn new concepts.
- You have some (pet) projects to review
Icing on the cake
- You have participated in an open data challenge (for example on Kaggle).
- You can deploy experimental models as a web application in python, for example using Docker, Flask and Keras.
- You have experience in Anomaly Detection and unsupervised learning.
- You have worked with industrial datasets before.
- You like automating and generalizing the boring, but important stuff (deployments, tests, workflow)