An open-source deep learning framework from Google, built around computation graphs, with Keras as its high-level API.
Its strength
A mature production and deployment story: TensorFlow Serving, TensorFlow Lite for mobile and edge, TensorFlow.js for the browser, and TensorBoard for visualization.
Where it runs on AWS
AWS Deep Learning AMIs and Containers — preinstalled with GPU drivers and CUDA.
Amazon SageMaker AI — managed training and hosting with a first-class TensorFlow estimator.
Accelerated by Trainium and Inferentia via the AWS Neuron SDK. See AWS ML Infrastructure.
Where it sits in the stack
Tier 3, ML Frameworks — you own the environment, scaling, and orchestration.
Not to be confused with
PyTorch — the other major framework, dynamic-graph and research-first. AWS supports both; pick on team familiarity, not AWS capability.