AWS Cloud Platform Architect: Zero-Downtime & Multi-Region

tonder; Ciudad de México; Remote; MXN 1,200,000 - 2,400,000; Full time

Job summary

tonder is seeking a Cloud Application Architect to design and own how our platform runs on AWS, from Docker Compose on a single host to ECS across multiple AZs, with Aurora PostgreSQL and zero-downtime deploys. You will work hands-on in Terraform and pipelines, partnering with an SRE team to operate what you design.

You will own runtime architecture for Django API and Celery, migrate the data tier, manage caches and messaging, and implement cost visibility and SRE practices across a

Qualifications

• 7+ years in engineering with 3+ years architecting production workloads on AWS
• Experience with ECS on EC2/Fargate or Kubernetes in production
• Terraform module level with CI workspaces and remote state
• Aurora/RDS PostgreSQL migrations with near-zero downtime
• Experience with PCI DSS or SOC 2 scope and audits

Responsibilities

• Design runtime architecture for Django API and Celery on ECS
• Migrate data tier to Aurora with near-zero downtime
• Manage cache and messaging with Redis/Valkey, SQS, Kinesis
• Plan and execute zero-downtime migrations and disaster recovery
• Implement IaC with Terraform modules and GitHub Actions pipelines
• Ensure observability with Datadog APM and SLO dashboards
• Lead on-call rotation and runbooks

Skills

AWS

ECS/Fargate/Kubernetes

Terraform

CI/CD

Datadog

Linux containers

GCP/Azure experience

Tools

Terraform

GitHub Actions

Datadog

Aurora PostgreSQL

SQS/Kinesis

Job description

tonder is seeking a Cloud Application Architect to design and own how our platform runs on AWS, from Docker Compose on a single host to ECS across multiple AZs, with Aurora PostgreSQL and zero-downtime deploys. You will work hands-on in Terraform and pipelines, partnering with an SRE team to operate what you design.

You will own runtime architecture for Django API and Celery, migrate the data tier, manage caches and messaging, and implement cost visibility and SRE practices across a