Use your extensive knowledge of TensorFlow, scikit-learn, and Python’s machine learning suite to help our data science team train machine learning models and develop insights on high-resolution time-series data. You will write feature extraction transformers, train models and tune hyperparameters using a visual analytics workflow to produce operational models that run on real-time data streams. We submit pull requests, do code reviews, write docstrings, and consider not only the accuracy of our models but also the efficiency of our code. This position reports directly to the Chief Data Scientist.
A Typical Week
- Perform analytics and data wrangling tasks on dense telemetry time-series data.
- Implement prototype models to test hypotheses related to event and anomaly detection and classification.
- Train and validate models using TensorFlow, tuning hyperparameters for optimal model performance both during training and inference.
- Develop visualizations to better interpret and understand model behavior.
- Brainstorm with team to conceptualize new features and hypotheses.
- Work with a range of external stakeholders to understand deliverables.
- Write and publish papers both for power engineering and machine learning conferences.
- Mentor other software engineers and review code.
- Masters or PhD in Statistics, Mathematics, Computer Science, Data Science, or another quantitative field.
- 5+ years of Machine Learning experience including both supervised and unsupervised methods, preferably across a variety of domains; experience with TensorFlow and scikit-learn is preferred.
- Expert use of the Pydata suite (numpy, scipy, pandas, matplotlib, etc.) to manipulate data and draw insights from large data sets.
- Knowledge of advanced statistical techniques and concepts (regression, properties of distributions, statistical tests and proper usage, etc.) and experience with applications.
- Understanding of time-series analytics and statistical methods.
- Excellent written and verbal communication skills for coordinating across teams.
- Experience participating in a software engineering team to produce operational machine learning models.
- A drive to learn and master new technologies and techniques.
- Experience in startup environments.
- Experience with time-series databases.
- Go programming.
- Distributed computing with Spark or other frameworks.
- Power and electrical engineering experience.