Staff Careers

The Research assistant will be responsible for completing an AFRL funded project related to identification of pervious and impervious regions in a city from aerial data. The selected incumbent will be required to work 10 hours per week for a period of three months, a duration which might be extended if required by the project. The work will involve using annotated aerial images for building a deep learning system for the binary classification problem of identifying pervious (water percolates) and impervious (water does not percolate) regions. The code will have to be shared with the PI through GitHub. The data will be provided by the PI and will be part of the project resources which will need to be handed back at the end of the employment term.

Duties/Responsibilities
• Conduct research focused on developing a deep learning system for the binary classification of land areas as either pervious or impervious.
• Perform research, coding, and testing the results.
• Utilizing annotated aerial images to support the project's objectives.
• Documenting all workflow steps in Github or other repositories.
• Work closely with the PI and other team members to align project goals and methodologies.

• College coursework, preferably in Science, Engineering, or Business
experience in working with aerial datasets related to the classification of pervious vs impervious regions.
• Knowledge of Python, PyTorch, Linux, Machine Learning, Deep Learning and Image Processing required.

• MS in Computer Science is strongly preferred

$37,440/2 =$18