Miramo

Miramo builds a French PropTech on Clever Cloud’s sovereign cloud

Miramo aims to simplify the rental search process. For landlords, the platform helps property owners find the right tenant. For tenants, it helps them find a rental home that genuinely matches their criteria.

Founded by Samuel Belolo, Founder & CTO, Miramo positions itself as an alternative to traditional property platforms. The goal is not simply to publish listings or collect applications. It is to build a clearer, more reliable and more human journey for both sides of the rental market.

Miramo is a matchmaking solution for the rental search process: on the landlord side, we find the right tenant; on the tenant side, we find the right flat. In short, the solution that gets everyone on the same page.

To support this ambition, Miramo chose Clever Cloud from the very beginning of the project. The decision was guided by three criteria: data sovereignty, ease of deployment and proximity with the support teams.

Rehumanising the rental search process

Miramo was born from a challenge familiar to many landlords and tenants. On the landlord side, the applications received do not always match the criteria set out in the listing. Some documents may be modified, fraudulent or difficult to verify. On the tenant side, the search often means applying in bulk to listings that are already unavailable, irrelevant or sometimes fake.

To address these issues, Miramo was built around four pillars: humanisation, simplification, security and personalisation. The platform is not trying to add yet another digital layer to the rental journey. It aims to offer a different path, where profiles are easier to understand, listings are more relevant and the relationship between landlords and applicants is clearer.

In a saturated market, very good profiles, real gems, end up lost in the crowd and go unnoticed. Putting the human being back at the centre of the process is essential, in a world where the rental search is becoming increasingly dehumanised.

This approach guides Miramo’s product decisions. It also shapes its technical choices, especially in the way the platform handles the personal information required for matchmaking.

Minimising data to reduce risk

Property rental involves personal information with significant stakes: identity, employment situation, income, supporting documents, search criteria and exchanges between landlords and tenants. For Miramo, this reality shaped product and technical decisions from the outset.

Our principle is minimisation: we only record what is strictly necessary.

The tenant verification journey is delegated to DossierFacile via OAuth. This French public digital service enables tenants to create a compliant, secure rental file labelled by the French State, without repeatedly sending supporting documents.

As a result, Miramo does not host identity documents, payslips or tax notices. The platform only retrieves the information needed for matching and connecting users. Access is controlled, retention periods are limited, and documents or images are isolated in dedicated object storage with restricted access.

This approach reduces data exposure by limiting what Miramo keeps in its own systems. It also aligns with the GDPR principle of data minimisation, which requires personal data to be adequate, relevant and limited to what is necessary for the intended purpose.

Choosing a cloud foundation aligned with this approach

Clever Cloud was already familiar to Samuel Belolo before Miramo was created. He had used it for several years on personal projects, particularly with a PHP stack.

When launching Miramo, the stack changed. Samuel therefore checked that the components required for the project were compatible and available. The choice of Clever Cloud was then confirmed around specific criteria: data sovereignty, ease of deployment and the relationship with support.

A privileged support relationship, ease of deployment and, above all, data sovereignty.

Before confirming this choice, Samuel also looked at other options on the market. The final decision remained pragmatic.

When I created the company, I chose a platform I already knew, a pragmatic choice that allowed me to move quickly from the start.

For Miramo, the infrastructure had to remain aligned with the way the product is designed: limiting the data stored, securing the exchanges needed for matchmaking and keeping operations easy to understand as the service evolves. In this context, Clever Cloud provides a French and European foundation, GDPR-compliant, ISO/IEC 27001:2022 certified, and simple enough to integrate without slowing down product development.

Deploying without carrying the operational burden

Miramo uses Clever Cloud for the main back-end components of its platform: NestJS API, workers, databases, Redis for cache and processing queues, S3-compatible object storage for images and documents, as well as staging and production environments.

This architecture fits into a straightforward delivery process. Each environment has its own branch. Changes go through a Pull Request, with tests, build and security audit before merge. Once the merge is complete, deployment is automatically triggered on Clever Cloud.

This workflow has remained stable since the initial implementation. For Samuel Belolo, this matters: the team has not had to reinvent its deployment process as the product has progressed.

Not having to manage infrastructure allowed us to focus all our efforts on the product.

At seed stage, this gain is concrete. Miramo has not needed to carry a dedicated DevOps or SRE role at this stage, which Samuel Belolo describes as the equivalent of avoiding a full-time hire. Production releases take a few minutes through a simple merge, with no manual server configuration.

Managed add-ons — database, Redis, object storage — also prevent the team from having to manage backups, patches or low-level monitoring themselves. At the start, Miramo estimates that it had an operational infrastructure in a few days, rather than several weeks.

That time could be reinvested in the priorities of the moment: making Miramo known, listening to landlords and tenants, and continuing to co-build the product with them.

Integrating AI within a controlled framework

As Miramo progresses, Samuel Belolo is also strengthening his internal AI tooling, notably with custom MCP servers. These tools make it possible to connect AI assistants to existing systems in order to query data, consult logs or automate certain checks.

In this context, Clever Cloud fits into Miramo’s day-to-day workflow without changing the principles established from the beginning. When their role is purely consultative, Miramo’s AI tools can be limited to read-only access. This separation adds a safeguard at infrastructure level, in addition to the rules defined in the code.

The Clever Cloud API also makes it easier to query the state of the platform from these internal tools: tracking ongoing deployments, consulting instance logs or quickly checking the behaviour of the API and workers.

The metrics available through Warp 10 complete this approach. They provide access to indicators such as memory and CPU usage, which are useful for detecting drift, anticipating scaling needs or spotting a memory leak before it becomes an incident.

Supporting Miramo’s growth

Miramo is now entering a phase in which the product needs to gain visibility, usage and maturity. The next challenges are scalability and the deployment of new services as the platform grows.

In this trajectory, Clever Cloud is not only a technical starting point. It is a foundation the team can continue to evolve with the product, without shifting its attention towards infrastructure administration.

For Samuel Belolo, this simplicity sums up Miramo’s experience with Clever Cloud.

The infrastructure that gets everyone on the same page.

Managed Kubernetes: benefits, limitations and selection criteria

A managed Kubernetes service is an offering in which the cloud provider manages the Kubernetes control plane on the user’s behalf. Provisioning, updates, availability and certificate rotation are handled by the provider. The user retains control over their workloads and node pools but does not administer the cluster’s critical infrastructure.
Engineering Features

Kubernetes Cloud: What It Is, How It Works, and the Main Types of Offerings

People often talk about “Kubernetes cloud” as if it were a single product, or set Kubernetes and the cloud against each other as competing choices. Neither is accurate.
Engineering

Clever Cloud’s partnership strategy: building an open ecosystem for strategic autonomy

When we launched our PaaS, Clever Cloud, our mission was clear: to create a cloud platform that developers would trust and enjoy using, something reliable, intuitive, and well-suited to the evolving demands of modern software development.
Company