Skip to content
Public template

About

No description, website, or topics provided.

Resources

Code of conduct

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Repository files navigation

CSAI 380: AI Operations — Block Project (STEDI)

Over the course you build one project in stages: you deploy the STEDI balance-device application to your own GKE Autopilot cluster, then add persistent storage, a batch ETL pipeline, event streaming with Kafka, and a scaled downstream integration. Each week's N.5 Block Project builds on the previous one, so complete them in order.

Each student works in their own Vocareum-provided GCP project and uses the GKE Autopilot cluster they created in 1.4 Build, Deploy, and Explore (region us-east1).

Final project architecture

Table of Contents

Week Assignment What you build Status
1 1.5 Block Project: Collecting the Data Containerize and deploy STEDI to GKE, Terraform namespace/registry, GitHub Actions build/push, architecture v1 ✅ VERIFIED 2026-10-30
2 2.5 Block Project: Saving the Data Redis and MySQL StatefulSets, headless services, secrets via secretKeyRef, persistence experiment ✅ VERIFIED 2026-10-30
3 3.5 Block Project: Working ETL Pipeline Airflow ETL DAG, pandas transform, idempotent load into MySQL, troubleshooting loop 🚧 DRAFT
4 4.5 Block Project: Streaming or Micro-batches Kafka via Helm, customer-risk topic, producer/consumer round trip, processing models ✅ VERIFIED 2026-10-30
5 5.5 Block Project: Scaling and Downstream Integration Consumer lag, scaling experiment, idempotent upsert, analytical SQL, final architecture 🚧 DRAFT

Each assignment's status banner sits directly under its title. ✅ VERIFIED means every step was run against a real GKE/GCP environment; 🚧 DRAFT means the instructions may still change.

Starter files

The files the assignments tell you to use, copy, or complete are in starter/:

Folder Contents First used
starter/terraform/ Namespace and Artifact Registry repository (later a GCS bucket) 1.5
starter/.github/workflows/ GitHub Actions workflow that builds, pushes, and deploys STEDI 1.5
starter/k8s/stedi/ STEDI Deployment and Service 1.5
starter/k8s/stateful/ Redis and MySQL StatefulSets, services, and secret example 2.5
starter/dags/, starter/data/, starter/helm/airflow-values.yaml Airflow ETL DAG, sample CSV, and Airflow Helm values 3.5
starter/helm/kafka-values.yaml, starter/k8s/kafka/ Kafka Helm values, client and consumer pods, STEDI Kafka patch 4.5
starter/app/ MySQL sink consumer 5.5

Values you must replace are written as <angle-brackets>, and code you must write is marked # TODO(student):.

Related

About

No description, website, or topics provided.

Resources

Code of conduct

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages