Exams / AIP-C01 / Study guide / Building GenAI front ends and business integrations (Amplify, OpenAPI, Bedrock Data Automation)
Building GenAI front ends and business integrations (Amplify, OpenAPI, Bedrock Data Automation): AIP-C01 study notes and practice questions
What to know about Building GenAI front ends and business integrations (Amplify, OpenAPI, Bedrock Data Automation) for the AIP-C01 exam, from the Implementation and Integration domain (26% of the exam guide), then 1 free question to test it. Part of the AIP-C01 study guide.
AIP-C01 asks for the quickest or lowest-effort way to deliver a generative AI interface or business workflow while keeping control of data and code. The answer is usually the managed service built for that layer.
Front ends with Amplify
AWS Amplify Hosting deploys a web front end from a Git repository, with CI/CD and preview environments for pull requests. The Amplify AI kit adds prebuilt chat components and conversation routes backed by Bedrock models. A team that needs a working chat interface with sign-in and hosting in days should start there instead of building the UI and authentication from scratch.
Contracts with OpenAPI
Write an OpenAPI specification before the back end is built. Front-end and back-end teams generate client code and mock servers from the same contract and work in parallel. API Gateway can import OpenAPI definitions, and AgentCore Gateway can expose OpenAPI-described APIs as MCP tools for agents.
Business workflows with Step Functions
A document flow might work like this: an S3 upload starts a Step Functions workflow, Bedrock Data Automation extracts fields, a Lambda function validates them, a task token pauses the workflow for human approval, and the result is stored in DynamoDB. Step Functions provides retries, error handling and an execution history that operations teams can inspect.
Field extraction with Bedrock Data Automation
Bedrock Data Automation extracts structured output from documents, images, audio and video. You define the fields in a blueprint instead of writing and tuning a prompt for each document type. It is the better fit for invoices, forms and claims than a general prompt.
No-code flows
Amazon Bedrock Flows provides a visual builder with prompt, knowledge base, Lambda, condition and iterator nodes. Business analysts can assemble a sequence of model steps without custom orchestration code. When the logic outgrows the visual builder, move it into Step Functions.
CRM enhancement
A CRM event can trigger a Lambda function that calls a model to write a call summary and then updates the CRM record through its API. Route the event with EventBridge or a CRM webhook, as covered in the enterprise integration topic.
How to choose
Scenario wording
Choose
“Chat front end with sign-in and hosting in days”
Amplify with the AI kit
“Several teams integrate before the back end is finished”
OpenAPI specification first
“Long document flow with retries and approval”
Step Functions workflow
“Pull named fields from invoices into a database”
Bedrock Data Automation blueprint
“Business analysts assemble model steps visually”
Amazon Bedrock Flows
“Summarize each CRM call record automatically”
CRM event to Lambda that calls the model
Common traps
Hand-building a chat UI and login when Amplify provides hosting and authentication.
A general prompt for form extraction when Bedrock Data Automation blueprints are built for it.
Bedrock Agents Classic for new action groups. It is closed to new customers, so use Amazon Bedrock AgentCore.
Amazon Q Business for a new assistant. It is closed to new customers, so use its successor, Amazon Quick.
Writing the API after the front end, which forces rework when the teams disagree on fields.
A regional credit union wants a chat assistant for its branch staff. The assistant answers questions from the union's policy handbook, and staff sign in with their work accounts. Replies should appear as they are written, and each employee should see only their own past conversations.
The union's IT team has two developers who know React well but have no designer and have never built a streaming chat backend. They need a working version within about two weeks, and the backend should call models on Amazon Bedrock.
Which approach meets these requirements with the LEAST development effort?
Choose one answer.
AA. Build the chat interface from scratch in React, with a custom WebSocket server on Amazon EC2 for streaming, and store each employee's history in a self-managed database.
BB. Build a Streamlit application that runs on an EC2 instance behind an Application Load Balancer, with a login page the team writes itself and a local file for each user's history.
CC. Publish a static HTML page that calls one Lambda function, which returns each answer from Amazon Bedrock, with no conversation history so that the page stays simple.
DD. Create an Amplify Gen 2 app with the Amplify AI kit, using its prebuilt React chat component, per-user conversation history and sign-in in one project.
Show the answer and explanation
Answer: D
A. Wrong: A custom interface, server and database would take far longer than two weeks for two developers. Each part is also one more component to secure and maintain.
B. Wrong: Streamlit gives a quick interface, but the team must write its own sign-in and manage the instance and history files. Local files are also awkward to keep safe for each employee.
C. Wrong: A static page with one call is quick to publish, but the union needs history and streaming, which this design leaves out. Staff would also lose their earlier questions when the page reloads.
D. Right: The AI kit supplies a React chat component, streaming responses from Amazon Bedrock and conversation history stored for each signed-in user, so the developers write little UI or backend code. Sign-in is part of the same Amplify project, which keeps the work within two weeks. Learn more: Amplify AI kit (opens in a new tab)
Explanation
The Amplify AI kit provides a React chat component, streaming responses from Amazon Bedrock and per-user conversation history, and Amplify Gen 2 adds sign-in in the same project. That gives the two developers a working assistant in about two weeks. Custom front ends, a self-written sign-in and a stateless page each take longer or miss a stated requirement.
Want more on Building GenAI front ends and business integrations (Amplify, OpenAPI, Bedrock Data Automation)? The 5 full-length practice sets have 5 more Building GenAI front ends and business integrations (Amplify, OpenAPI, Bedrock Data Automation) questions, 375 in all, none of them in the free exam. One payment of $14.99, no subscription. See the practice sets →
Study Building GenAI front ends and business integrations (Amplify, OpenAPI, Bedrock Data Automation)