Pro
Paul G
Expert cloud & platform consultant building AI platforms
New to Contra
Paul is ready for their next project!
Here's how I'd build a scalable ticketing backend with Supabase & Stripe.
The backend is where everything lives or breaks. Supabase handles auth and Row Level Security out of the box, Postgres enforces ticket inventory and access at the database level, and Stripe Checkout processes payments with webhooks that keep everything in sync. Capacity limits, discount logic, and user access all of it stay consistent when real traffic hits. Built to scale without needing a full rewrite six months down the road.
0
15
Cross-cloud real-time data pipeline for an e-commerce platform processing 2M+ events per day. Event sources feed into SQS queues and AWS EventBridge. Lambda functions handle initial transformation and validation before landing raw data in S3. A cross-cloud sync layer pushes events into GCP Pub/Sub for real-time stream processing. Dataflow streaming jobs handle enrichment, deduplication, and windowed aggregations before passing to Cloud Run microservices for feature engineering. Vertex AI runs real-time prediction models for recommendation scoring and fraud detection. Processed features land in Redis for low-latency serving and BigQuery for analytics. Cloud Storage holds Parquet files for batch retraining. Cloud Build orchestrates the ML pipeline, including model training, evaluation, and deployment.
1
47
Architected an enterprise Kubernetes platform on Azure AKS for a fintech company running 40+ microservices. Traffic enters through Application Gateway with WAF v2 policies and Azure Firewall before reaching the AKS cluster inside a hub-spoke VNet. ArgoCD manages all deployments through GitOps with automatic drift detection and self-healing. ACR stores container images built by Azure DevOps Pipelines with integrated vulnerability scanning. Workloads connect to Cosmos DB, PostgreSQL Flexible Server, and Redis Cache through private endpoints. Azure ML handles fraud detection model inference called directly from within cluster pods. Entra ID with Managed Identities provides passwordless auth across all services.
0
24
Designed and deployed a multi-account AWS landing zone using Control Tower and Organizations for a healthcare SaaS company handling PHI data. The architecture separates management, security, network, and workload accounts with dedicated IAM boundaries. Traffic routes through CloudFront and WAF into a Transit Gateway hub before reaching production VPCs running ECS Fargate workloads. Security Hub aggregates findings from GuardDuty, Inspector, and AWS Config across all accounts. CloudTrail feeds into EventBridge for real-time alerting through SNS. All encryption managed through a centralized KMS account. Infrastructure provisioned entirely with Terraform and Terragrunt modules. Achieved SOC 2 Type II and HIPAA compliance within 90 days.
0
36
I built a production-ready serverless LLM API on GCP designed for low cost, strong security, and fast inference. Requests flow through CDN, load balancing, WAF, and API management before hitting a Cloud Run FastAPI service that handles prompts, session history, caching, and model routing. The system switches between Gemini 2.5 Pro for deep reasoning and Gemini Flash for fast responses, with RAG support using Vector Search over 768-dim embeddings. Data is stored in Firestore, cached in Redis, and logged to BigQuery. Everything is secured with VPC Service Controls, Workload Identity, KMS, Secret Manager, and DLP. CI/CD is fully automated with Terraform and Cloud Build using canary rollouts and auto-rollback on SLO violations. At around 50K requests per day, the platform runs at about $1K/month and scales to zero when idle.
0
50