How does AI work in the cloud? My friend Lynn encouraged me to learn about bioinformatics, the convergence of Software (informatics) and Life Sciences (bio). Machine learning and AI are playing fascinating roles such as identifying protein structure from images.
One of the field’s biggest challenge is: scale. There’s just so… much… data. This makes cloud scalability very attractive: it lets us focus on the actual task rather than purchasing hardware and installing operating systems.
Here’s a demo (+ source code!) for a “Hello World” machine learning application to recognize hand-written digits:
- train a neural network on the MNIST dataset using Python + TensorFlow on Google Vertex AI
- serve the trained model via Vertex AI Endpoints
- wrap the model in a Cloud Function that translates attached images into pixel vectors for the AI
Source + instructions available on Github: dchaley/hello-mnist