01
case 01 / 05
Nexus AI Content Engine
Enterprise-grade serverless platform for automated marketing content generation and distribution.
Year
2025
Role
Cloud architect · Full-stack
Status
Proprietary · internal infrastructure
( architecture )
How it's built
AWS LambdaStep FunctionsS3EventBridgeOpenAINext.js
Hover a service to trace its connections
ClientEdge / APIComputeEvents / queuesStorageAI / ML
( the problem )
Marketing teams were producing copy and creatives by hand for every channel, so a single campaign took days and never looked consistent across networks.
( the build )
A fan-out pipeline orchestrated by Step Functions: one brief becomes channel-specific copy, generated images and a publishing schedule, all event-driven and paid per execution.
38s
brief → full campaign
12×
more assets per week
$0
idle infrastructure cost
( request flow )
- 01A brief is submitted from the Next.js studio and lands on API Gateway.
- 02An orchestrator Lambda starts a Step Functions state machine per campaign.
- 03Parallel branches generate copy with OpenAI and render images into S3.
- 04EventBridge schedules each publication and a publisher Lambda ships it to every network.
nexus.console● live
38s
brief → full campaign
12×
more assets per week
$0
idle infrastructure cost
handler.ts
1const run = await sfn.startExecution({
2 stateMachineArn: NEXUS_PIPELINE,
3 input: JSON.stringify(brief),
4});
5// fan-out → copy · images · schedule
6▍
main ✓TypeScript · UTF-8 · Ln 6
Next case · 02
Sentinel Guard →