Millions of Indonesians need a safe place to talk about how they feel, yet many never reach one. Temucita is changing that with community-based support spaces and an AI emotional companion. As the platform grew, it needed cloud infrastructure that was as secure and dependable as the care it offers. CASA5 led Temucita’s GCP to AWS migration, moving its workloads onto a governed, well-secured AWS foundation in Jakarta.

About Temucita: recover, connect, grow

Founded in 2025, Temucita is an Indonesian mental wellness platform built around its promise: “Pulih. Terhubung. Bertumbuh.” (Recover. Connect. Grow.). It is not a clinic or a therapy app. It is a community where people find support in three ways:

• Safe Sharing Sessions: facilitated peer-support groups on many topics.
• AI Emotional Companion: private, on-demand emotional support, whenever someone needs it.
• Emotion regulation tools: guided exercises for handling difficult feelings.

According to Temucita, the platform has already brought together 30+ professionals and 50+ safe spaces across Greater Jakarta, has reached 2,000+ people, and reports 98% satisfaction.

The challenge: growing fast on a foundation that could not keep up

Temucita originally ran on Google Cloud Platform, with containerized services on Cloud Run and a PostgreSQL database on Cloud SQL. That was a sensible way to launch. But to scale a platform that handles people’s emotional conversations, the team needed more:

• Stronger governance and security for sensitive user data, with separate environments for development, staging and production.
• A foundation ready for growth, including a future multi-tenant SaaS model for partners and organizations.
• Better long-term cost control through managed services and right-sized capacity.
• Access to the AWS generative AI ecosystem, including Amazon Bedrock, to take its AI companion further.
• Automation: there was no CI/CD pipeline yet, so infrastructure changes were manual.

Temucita decided to move to AWS and chose CASA5 to plan and deliver the migration.

The solution: a secure AWS landing zone and a clean migration

CASA5 approached the project in two layers: first build a well-governed home on AWS, then move the applications and data into it. Work started with a kickoff and workload discovery in mid-June 2026. The core migration was completed in early July.

1. A multi-account landing zone with AWS Control Tower

• AWS Control Tower set up a multi-account organization, with dedicated organizational units for security, infrastructure, non-production and production workloads, plus central Log Archive and Audit accounts.
• Account Factory for Terraform (AFT) automates the creation of new AWS accounts, so every new environment follows the same rules.
• Separate development, staging and production accounts keep environments isolated, so a mistake in testing cannot affect real users.
• AWS IAM Identity Center is connected to Temucita’s Google Workspace, giving the team single sign-on (SSO) to every account with one set of credentials.

2. A security baseline across every account

Security was designed in from day one, not added later. The landing zone includes AWS CloudTrail for audit logging, AWS Config and AWS Security Hub for continuous compliance checks, and Control Tower guardrails with Service Control Policies to prevent risky configurations. Logs from every account flow into a central, protected log archive.

3. Everything as code with Terraform and GitHub Actions

All AWS infrastructure is defined in Terraform with centralized remote state, and changes go through a new GitHub Actions CI/CD pipeline. No more clicking through consoles: every change is reviewed, repeatable and documented.

4. Applications: from Cloud Run to Amazon ECS on Fargate

Temucita’s containerized services moved from Google Cloud Run to Amazon ECS on AWS Fargate, with images stored in Amazon ECR. An Application Load Balancer with an SSL certificate from AWS Certificate Manager routes traffic securely. The production network runs in a dedicated Amazon VPC spread across two Availability Zones, with separate public, private and database subnets.

5. Data: from Cloud SQL to Aurora PostgreSQL

The PostgreSQL database moved from Google Cloud SQL to Amazon Aurora PostgreSQL for production and Amazon RDS for PostgreSQL for non-production. Using a planned dump-and-restore cutover, the data was moved in one controlled step during an agreed maintenance window.

The whole platform now runs in the AWS Asia Pacific (Jakarta) Region, so Indonesian users are served from close by and their data stays in Indonesia.

The results: a platform ready to care for more people

• A trusted home for sensitive conversations. Environment isolation, central audit logs and continuous compliance checks give Temucita a strong base for protecting users’ emotional data.
• Ready to scale. Serverless containers on Fargate grow with demand, and new accounts or environments can be created automatically for new products or partners.
• Faster, safer releases. Infrastructure as code and a CI/CD pipeline let a small team ship changes with confidence.
• Lower operational overhead. Managed services such as Fargate and Aurora mean the team spends its time on the product, not on servers.
• A team that owns its cloud. CASA5 handed over as-built documentation, operational runbooks and knowledge transfer, so Temucita can run and extend the environment independently.

What’s next: generative AI for mental wellness on Amazon Bedrock

Moving to AWS also opens the door to Temucita’s next chapter. In an earlier assessment, CASA5 and Temucita identified two high-impact generative AI opportunities built on Amazon Bedrock:

• A more empathetic AI companion that handles rich, multi-turn conversations using foundation models on Amazon Bedrock.
• A copilot for psychology professionals that uses Knowledge Bases for Amazon Bedrock to draw on anonymized case studies and Cognitive Behavioral Therapy (CBT) materials, helping professionals prepare and support people more effectively.

With a secure, governed foundation already in place, Temucita can build these capabilities on solid ground.

Why CASA5

CASA5 is an AWS partner based in Jakarta that helps Indonesian companies plan, migrate and build on AWS, from landing zones and cloud migrations to generative AI. For Temucita, we combined AWS best practices, infrastructure as code and a clear handover so a fast-growing startup could get enterprise-grade infrastructure without an enterprise-sized team.

Frequently asked questions

Why did Temucita migrate from GCP to AWS?

Temucita wanted a scalable multi-account governance foundation, stronger security for sensitive user data, long-term cost optimization through managed services, and access to the AWS partner and generative AI ecosystem, including Amazon Bedrock.

What is an AWS landing zone?

An AWS landing zone is a pre-configured, secure, multi-account AWS environment. It sets up accounts, identity, logging and security guardrails up front, so every workload starts from the same well-governed baseline.

Which AWS services replaced Google Cloud Run and Cloud SQL?

Google Cloud Run was replaced by Amazon ECS on AWS Fargate, and Google Cloud SQL was replaced by Amazon Aurora PostgreSQL for production and Amazon RDS for PostgreSQL for non-production.

Where is Temucita’s data hosted after the migration?

Temucita’s platform runs in the AWS Asia Pacific (Jakarta) Region, so it serves Indonesian users from close by and keeps their data in Indonesia.

Planning your move to AWS?

Whether you are running on another cloud or on your own servers, CASA5 can help you migrate to a secure, well-governed AWS environment and prepare for what comes next, including generative AI.