Module 2 Prompt Engineering and Amazon Q Flashcards
Amazon Q Business
Fully managed Gen-AI assistant for employees based on a company’s internal knowledge and data
RAG
Retrieval-augmented generation; combines model capability with external data sources for more informed responses
IAM Identity Center
Authentication system used to control access to Amazon Q Business
Admin controls
Settings to customize responses in Amazon Q Business; similar to guardrails in Amazon Bedrock
Amazon Q Apps
Feature to create Gen AI-powered apps without coding using natural language
Amazon Q Developer
Service that answers questions about AWS documentation and resources in AWS accounts
AI code companion
Feature of Amazon Q Developer to assist with coding AWS-based applications
PartyRock
Playground for building Gen AI apps powered by Amazon Bedrock; requires no AWS account
Amazon Q for QuickSight
Integration allowing natural language queries to generate data visualizations in QuickSight
Amazon Q for EC2
Feature to help choose appropriate EC2 instance types based on workload requirements
Amazon Q for AWS Chatbot
Integration allowing access to Amazon Q through AWS Chatbot in applications like Slack or Microsoft Teams
Prompt Engineering
The practice of developing; designing; and optimizing prompts to ensure foundation model output fits specific needs
Zero-shot prompting
Presenting a task to a model without providing any examples or explicit training for that specific task
Few-shot prompting
Providing examples of a task to the model to guide its outputs
Chain of thought prompting
Dividing a task into a sequence of reasoning steps leading to more structure and coherence
