This intern candidate will work closely with our co-founder/CTO and data science team on implementing bleeding-edge machine learning methods into our platform. The ideal candidate should have a strong academic background in ML and excellent programming abilities, preferably in Python.
Potential projects include:
1. Investigating deep learning network architectures for time-series survival modelling of industrial assets.
2. Experimenting with natural language processing (NLP) techniques using large quantities of unstructured text data from industrial maintenance and inspection records and integrating these techniques in our survival model pipeline.
3. Improving and building out our novel semi-supervised machine learning methods in the context of anomaly detection and operating mode classification.
4. Using AWS Kinesis streams to implement large-scale real-time anomaly detection for streaming time-series data.
- At least 1 year of academic or professional data science work or experience
- Deep experience with Python
- Experience working with very large datasets, especially using Dask and Kinesis
- Background in time series analysis, survival modeling, hidden Markov models, Gaussian processes, and variational inference.
Work with the founding team to solve some of the hardest problems in heavy industry (using lots of unique data!). We work with energy companies and utilities around the world to increase infrastructure reliability, reduce costs, and improve safety.