AI-900 and AI1-C01 Comparison
Personal Reflection
Azure segmentation is simpler (Azure AI Services and Azure OpenAI vs AWS SageMaker, Bedrock, all the Foundation Models and all the older AWS AI services.)
The AI services also seemed more fragmented in AWS, it took me longer to differentiate all of them in my head (eg Rekognition, Textract, and Comprehend).
I felt there was deeper theory on the AWS exam, such as NLP evaluation metrics (eg. BLEU, ROUGE).
So overall, AWS difficulty was higher due to complexity and depth, though some of this may be due to it being a Beta.
AI Workloads and Considerations
AI-900
Understanding AI workloads, ethical considerations, and responsible AI.
AI1-C01
Fundamental concepts and terminologies of AI, ML, and generative AI, including responsible AI.
Machine Learning on Azure
AI-900
Principles of machine learning, including supervised and unsupervised learning, and Azure Machine Learning services.
AI1-C01
Model training and fine-tuning, feature engineering, and AWS SageMaker for building, training, and deploying ML models.
Computer Vision Workloads
AI-900
Features of computer vision workloads, including image classification, object detection, and Azure Cognitive Services.
AI1-C01
AWS Rekognition for image and video analysis, and other computer vision services.
Natural Language Processing (NLP) Workloads
AI-900
Features of NLP workloads, including text analytics, language understanding, and Azure Cognitive Services.
AI1-C01
AWS Comprehend for NLP tasks, including sentiment analysis, entity recognition, and language detection.
Generative AI Workloads
AI-900
Features of generative AI workloads, including Azure OpenAI Service.
AI1-C01
Generative AI concepts and use cases, including AWS services for generative AI.