A large e-commerce company has a vast catalog of product images and needs to classify...

AutoML is appropriate because the company possesses a largely labeled image dataset and wants to train a customized image-classification model. AutoML simplifies model development by automating important activities such as architecture selection, training, tuning, and evaluation, reducing the amount of specialist machine learning code required. The labeled product images provide the input examples and target categories needed for supervised image classification. Gemini Code Assist supports software development and does not train image-recognition models. Google AI Studio is primarily used to prototype applications and prompts with Gemini models rather than create a dedicated supervised classifier from a labeled image collection. Agent Search focuses on retrieving enterprise information and is unrelated to image-model training. Therefore, AutoML on Gemini Enterprise Agent Platform is the offering that best satisfies the custom image-recognition requirement.