Writing the Python SDK
Joining the Edge Impulse SDK team presented an exciting opportunity to contribute to the development of embedded machine learning solutions.
I was tasked with developing a Python SDK to interface with popular data processing libraries like Pandas and NumPy. This SDK streamlined the process of uploading data and deploying models to the edge, making it accessible to Python developers.
Technology
Working on the Edge Impulse SDK team exposed me to various technologies, including data manipulation tools like Pandas and deep learning frameworks such as Keras, Scikit-learn and PyTorch
About the client
Edge Impulse builds tooling for machine learning on small devices: collect data, train a model, run it on a microcontroller in the field. Their users range from hobbyists to industrial fleets, which makes an SDK’s surface area an interesting design problem.
My experience with the Edge Impulse SDK team was enriching, as I collaborated with talented individuals and contributed to a project with significant potential in the embedded AI space. As we continue to refine our tools and empower developers, I am excited to see the impact of our work on the broader ecosystem.