Browse AWS Certification Guides

DVA-C02 Development with AWS Services Guide

Study DVA-C02 Development with AWS Services: key concepts, common traps, and exam decision cues.

This is the biggest DVA-C02 domain because AWS expects developers to classify application behavior before they worry about deployment polish. Questions here usually ask whether the stronger answer is a messaging pattern, a Lambda behavior change, a DynamoDB access-pattern fix, or a cleaner SDK integration.

I verified the current DVA-C02 Domain 1 task statements on May 11, 2026. AWS currently expresses this domain as three task statements:

  1. develop code for applications hosted on AWS
  2. develop code for AWS Lambda
  3. use data stores in application development

Current weight in the exam guide

AWS currently weights Development with AWS Services at 32% of scored content.

What this domain is really testing

This domain is not testing whether you can name every AWS service in the developer ecosystem. It is testing whether you can:

  • choose the right interaction pattern before you write code
  • understand how Lambda triggers, retries, configuration, and VPC access affect application behavior
  • recognize when a data-access problem is really a schema, key-design, consistency, or caching problem
  • prefer managed integrations that reduce glue code, duplicate side effects, and retry pain

Work this domain in order

Lesson AWS task fit Why it matters
1.1 Application Patterns, APIs, Messaging & Integrations Task 1.1 This is the application-shape lane: sync vs async, stateful vs stateless, APIs, messaging, SDKs, streaming, and resilient third-party integration code.
1.2 Lambda Config, Events, VPC Access & Errors Task 1.2 This is the Lambda behavior lane: parameters, event lifecycle, retries, DLQs, destinations, private access, integration, and performance tuning.
1.3 DynamoDB, Consistency, Caching & Data Stores Task 1.3 This is the data-access lane: key design, query vs scan, consistency, serialization, lifecycle controls, caching, and specialized stores.

Fast routing inside this chapter

If the question is really about… Go first to…
decoupling, event buses, queues, choreography, orchestration, APIs, SDK calls, or resilient third-party integrations 1.1 Application Patterns, APIs, Messaging & Integrations
timeouts, concurrency, handlers, layers, triggers, destinations, DLQs, private resources, or Lambda tuning 1.2 Lambda Config, Events, VPC Access & Errors
partition keys, query versus scan, consistency, indexes, serialization, caching, lifecycle controls, or specialized stores 1.3 DynamoDB, Consistency, Caching & Data Stores

If you keep missing questions in this domain

Symptom What is usually going wrong Fix first
every answer choice looks like a plausible service you are choosing a product before classifying the interaction pattern go back to 1.1 and separate synchronous request/response, buffered async, event fan-out, and orchestration
Lambda questions feel random you are not anchoring on trigger semantics, retry behavior, and timeout or concurrency together rework 1.2 and treat invocation model as the first fork
DynamoDB questions feel like memorization you are thinking in tables, not access patterns rework 1.3 and force every scenario into key design, read or write shape, and consistency needs
you keep overengineering you are drifting into architect mode instead of developer mode prefer the answer that fixes app behavior with the least infrastructure churn

What strong answers usually do

  • separate application pattern questions from deployment questions
  • prefer loose coupling and async buffering when the stem mentions spikes or downstream instability
  • recognize that Lambda reliability depends on configuration, retry behavior, and event-source semantics together
  • pick the data store that matches the access pattern instead of forcing every requirement into one familiar service

Common DVA-C02 traps in this domain

  • answering with a service name before classifying the interaction pattern
  • confusing query with scan, or strongly consistent with eventually consistent
  • assuming every Lambda integration should stay synchronous
  • treating resilience as only a deployment concern instead of a coding and integration concern

Before you leave this domain

Make sure you can do three things quickly:

  1. classify the flow as synchronous, asynchronous, event-driven, or stateful orchestration
  2. explain how failure and retry work for the chosen trigger or integration
  3. justify why the selected data store or cache matches the access pattern better than the distractors

Then move to 2. Security, because DVA-C02 often turns a good application-design answer into a wrong final choice by tightening the credential, encryption, or tenant-isolation constraint.

In this section

Revised on Monday, June 15, 2026
2. Security