AWS Serverless Metadata Workflow

Inspectable project files

Project evidence

The implementation can be checked through its source code, infrastructure template, test suite, deployment instructions, and operating documentation.

Implementation

The public repository contains a complete event-processing path rather than a frontend mockup.

Python 3.12

Lambda function

Validates S3 events, decodes object keys, calls HeadObject, normalizes metadata, and writes an idempotent DynamoDB record.

Open code

AWS SAM

Infrastructure

Defines S3, Lambda, DynamoDB, SQS, IAM permissions, encryption, retry settings, log retention, and CloudWatch alarms.

Open template

Pytest

Test suite

Covers normal processing, duplicate events, malformed events, versioned objects, access failures, and client reuse.

Open tests

Verification results

The August 3, 2026 local audit ran the same implementation and tests stored in the repository.

11passing unit tests
100%statement coverage
100%branch coverage
85%minimum CI coverage requirement

Automated checks

  • Python compilation and Ruff linting
  • Pytest with branch coverage
  • Static-site link, metadata, JSON-LD, XML, and asset validation
  • AWS SAM template validation and application build
  • Scheduled checks of the published GitHub Pages routes and assets

Documentation

The project includes the information needed to deploy, inspect, operate, troubleshoot, and remove the stack responsibly.

Deployment

Prerequisites, parameters, deployment commands, output discovery, and smoke testing.

Read deployment guide

Operations

Log review, alarm response, failure-queue investigation, replay controls, and retained-resource cleanup.

Read operations runbook

Data contract

The DynamoDB fields, required values, optional values, deterministic identity, and schema version.

Read data contract

Troubleshooting

Common permission, event, timeout, duplicate, deployment, and retained-resource failure modes.

Read troubleshooting guide

Scope of the evidence

The original internship capstone was built in isolated training and project environments. The public repository demonstrates the architecture and later engineering work without exposing internal Amazon material, customer data, or production-system access.