DVA-C03 study plan: pass in 4 weeks

This four-week plan for the AWS Developer - Associate exam allocates about 30 hours in proportion to the domain weights in the DVA-C03 exam guide: Development with AWS Services 30%, Security 26%, Testing and Deployment 22% and Troubleshooting and Optimization 22%. The outgoing DVA-C02 guide weighted them 32/26/24/18, so the main shift is more time on troubleshooting. The topics inside each week follow the task statements in the DVA-C03 exam guide (checked 30 October 2026).

Assume about 7 to 8 hours a week and an AWS account you can use. AWS recommends one or more years of hands-on development experience; a plan this short suits someone who already builds on AWS. If you are newer, stretch each week to two.

The plan at a glance

Each domain also has an explanation page: development, security, testing and deployment and troubleshooting and optimization. Keep about an hour of each week’s budget for questions on that week’s domain.

Week 1: development with AWS services (about 9 hours)

Build rather than read. AWS consolidates four DynamoDB skills into one, so core behaviour matters more than corner cases. Task 1.1 also lists AI-assisted code generation and review and integrating managed AI services, and Task 1.2 covers Lambda in a VPC, layers and extensions.

• Compute and APIs: Lambda (event sources, concurrency, timeouts, environment configuration), API Gateway integrations and authorisers.
• Data: DynamoDB keys, secondary indexes, read and write capacity, conditional writes and consistency. Compare with S3 and relational choices.
• Messaging and events: SQS, SNS, EventBridge, Step Functions. Know when each fits and how failures are handled (retries, dead-letter queues, idempotency).
• SDK and CLI use: pagination, retries with backoff, error handling, serialising data, streaming with Kinesis and Amazon Data Firehose.
• Data lifecycle and caching: S3 Lifecycle rules, DynamoDB TTL, choosing between ElastiCache, DynamoDB, S3 and RDS by access pattern.
• AI in development: use an AI coding tool for a feature, review what it generates, and call a managed AI service such as Amazon Bedrock from code.

Exercise: build an upload-triggered workflow where a file lands in S3, a function processes it and results land in DynamoDB, with failures going to a queue.

Week 2: security, including AI security (about 8 hours)

Security is the second-largest domain. AWS consolidates the encryption and sensitive-data skills from 13 skills across 2 tasks into 7 skills in 1 task, and a new Task 2.3 covers AI security.

• Identity: IAM roles and policies, role assumption, temporary credentials, resource policies. See AWS IAM basics.
• Data protection: KMS keys and grants, envelope encryption, encryption at rest and in transit, Secrets Manager and Parameter Store.
• Authentication for apps: Cognito user and identity pools, identity federation, presigned URLs for S3 and CloudFront, and how tokens reach your API.
• Certificates and sensitive data: AWS Certificate Manager, AWS Private CA, masking PII and PHI.
• AI security (Task 2.3): the guide lists requesting and managing access to AI services, data privacy controls (VPC endpoints, keeping inputs and outputs out of model training), filtering model inputs and outputs (content filtering, sensitive information detection, PII/PHI redaction, denied topics, prompt injection protection), AI agent interaction security (tool-use authorisation, session isolation, human approval) and keeping sensitive content out of monitoring logs. Read the Amazon Bedrock, Bedrock AgentCore and AWS PrivateLink documentation for these.

Exercise: rewrite a role that has wildcard permissions into one with only the actions and resources it needs, then prove it by testing a denied call.

Week 3: testing and deployment (about 6.5 hours)

• Pipelines: CodeBuild, CodePipeline, CodeDeploy, and deployment strategies (all-at-once, rolling, blue/green, canary). See CI/CD explained.
• Infrastructure as code: CloudFormation, AWS SAM and the CDK at a reading level; packaging and parameterising a template.
• Configuration: AppConfig, Parameter Store, Secrets Manager, environment variables, stage variables, versions and aliases.
• Testing: SAM local testing, JSON test events for Lambda and API Gateway, mock APIs, test environments (aliases, image tags, Amplify branches).
• Containers: the guide adds container skills: images in Amazon ECR with tagging and lifecycle policies, deployment on Amazon ECS task definitions, EKS manifests and AWS Fargate. Build an image, push it to ECR and run it on ECS with Fargate so you understand the workflow end to end.
• AI-assisted development: the guide lists AI tools for generating tests, regression testing and CI/CD support such as automated deployment approvals and post-deployment validation. Use an AI coding assistant on your week-3 project and review its output critically for correctness and security.

Exercise: ship a change through a pipeline with a rollback you have triggered on purpose.

Week 4: troubleshooting and optimization (about 6.5 hours)

• Observability: CloudWatch metrics, logs, alarms and Logs Insights, X-Ray tracing, structured logging, custom metrics, health checks with liveness and readiness probes, alerts on quota limits and deployment completion.
• Troubleshooting: throttling, cold starts, timeouts, permission errors, a message being processed twice.
• Optimisation: caching (CloudFront, ElastiCache, DAX, API caching), memory and concurrency tuning (provisioned concurrency, connection and thread pools), SQS and SNS messaging, query and index design.

This domain is 22% of the exam, up from 18% in the outgoing guide, and the guide adds AI tools for error analysis and optimisation recommendations to it. Spend the second half of the week on one timed run: 65 questions in 130 minutes is exactly two minutes each. Start with the sample questions, then review every miss by reading the AWS documentation behind it. See how to study for certification exams for review habits.

Before the exam

• Compare the exam guide task statements with the weeks above and move hours toward any skill you cannot do yet.
• Re-read the exam format so the question types and timing are familiar.
• Use the study resources page for free official material.

What will change in this plan

The domain weights are AWS’s own, so the hour split stands. AWS publishes guide revisions at least one month before they reach the exam, so recheck the guide’s revisions section before you book. The core services above underpin both the outgoing and the updated exam.