You will be working with a startup developing real-time prediction algorithms on manufacturing production-line data.
You will be responsible for using machine learning algorithms to develop real-time “big data” predictive models.
You will need to be able to collaborate very closely with other team members on a frequent basis.
Responsibilities:
- Using cutting-edge machine learning algorithms to develop real-time predictive
models on big manufacturing data sets.
- Development of effective data visualizations to communicate data features
Ideal Requirements:
- Experience with machine learning techniques including stochastic gradient descent
and regularized regression (LASSO, Ridge, Elastic-Net)
- Experience with dimensionality reduction techniques such as PCA, SVD,
dictionary learning
- Knowledge of basic linear algebra (eigenvectors, eigenvalues)
- Experience with data visualization techniques
- Experience with python development, including features such as multiprocessing,
shared memory, and unit tests
- Experience with development using python packages such as numpy, scipy, pandas,
matplotlib, scikit-learn, and pymongo
- Experience with database systems such as MongoDB and Redis
- Experience with asynchronous web communication with packages such as Socket.IO
- Experience with Git and collaborative coding
- Experience with Ubuntu/EC2 development environment
- (Nice to have) experience with Hadoop/Map-Reduce development
- Previous experience developing predictive models
- Previous experience developing data visualizations
If interested, please reach out at bryton [dot] shang [at] skedastic [dot] com!
You will be working with a startup developing real-time prediction algorithms on manufacturing production-line data. You will be responsible for using machine learning algorithms to develop real-time “big data” predictive models.
You will need to be able to collaborate very closely with other team members on a frequent basis.
Responsibilities:
Ideal Requirements: If interested, please reach out at bryton [dot] shang [at] skedastic [dot] com!