How Clikma works under the hood
For IT and architecture teams evaluating a learning platform: what components Clikma has, where it runs, how it integrates, how it scales, how it is backed up and how changes are released.
- Independent services for each business domain
- Kubernetes in the Santiago de Chile region
- API, SFTP and single sign-on
- Daily backups and 14-day recovery
Updated 2026-10-24
How does the Clikma platform work under the hood?
Clikma is a multi-tenant web platform (SaaS) made up of 12 independent business services, such as identity and access, training, e-learning, employee records, performance, documents, job profiles, indicators and the AI assistant, plus a web application and its own content players. It runs in containers on managed Kubernetes in Google Cloud, in the Santiago de Chile region, with a regional PostgreSQL database that keeps a standby replica in another zone.
Each client company has its own subdomain, branding and configuration on the same platform, without separate installations, and all of them receive the same updates. Integrations come in through the API with per-client credentials, through automatic uploads over SFTP or through single sign-on with SAML 2.0, LDAP, Microsoft Entra ID, Google or OAuth.
Operations are monitored continuously, the database is backed up every day with point-in-time recovery for the last 14 days, every change goes through code review and static analysis before it reaches production, and guaranteed availability is 99.5%.
- 12
- independent business services
- 4
- separate environments: development, demo, user acceptance testing and production
- 14 days
- of point-in-time database recovery
- 99.5%
- guaranteed availability
The platform's layers
From the web entry point to the data, each layer has a clear function and a well-known technology, maintained by the Clikma team.
The services communicate through APIs over HTTPS, and each one owns its database within the same regional instance. That way, a failure or an update in one service does not stop the others.
| Layer | What it does | How it is built |
|---|---|---|
| Web entry point | Receives all HTTPS traffic, routes each subdomain to the right application and renews certificates automatically. | Gateway on Kubernetes, with automatic TLS certificates. |
| Web application | The interface for employees, managers and administrators, in Spanish, English and Portuguese, with WCAG 2.1 and 2.2 Level AA accessibility. | Angular application. |
| Content players | Play SCORM 1.2 and 2004 packages, xAPI, LTI, H5P, video and interactive HTML, with each company's branding. | Clikma's own players. |
| Business services | Identity and access, training, e-learning, employee records, performance, documents, job profiles, indicators, AI assistant and support services. | 12 services in Java with Spring Boot, one per domain. |
| Data | Each service's information, user sessions and files. | Regional PostgreSQL with a standby replica, high-availability Redis for sessions, and object storage. |
| Artificial intelligence | Writes the assistant's answers and searches the catalog and lessons by meaning. | Local language model hosted on Clikma's infrastructure, and a vector index. |
Multi-tenant: one platform, each company with its own setup
Clikma is multi-tenant by design. Each company's data is logically separated within the platform, and each user accesses only what their company and their privileges allow.
Multi-tenant LMS with custom branding and domain
Color palette, logo, loading screen and subdomain for each client, without separate installations.
Custom roles and permissions
The company administrator builds their own roles from the base roles.
Tailored organizational structure
Hierarchies, dynamic groups and attributes specific to each company.
Integrations and API
Clikma connects with what the company already uses: the HR system or ERP, the identity provider, video conferencing and analytics tools.
Integration with the HR system (HRIS)
The employee roster is synced with the company's HR system or ERP.
API with per-client credentials
Dedicated credentials, an IP allowlist and privileges defined for each company.
Automatic data upload over SFTP
File readers configurable for each client, and a history of every upload.
Single sign-on (SSO) with SAML 2.0 and LDAP
With the company's identity provider.
API access to the metrics
To bring Command Center indicators into other systems.
Webinars and video conferencing
The invitation includes the Zoom, Microsoft Teams or Google Meet link.
Fixed outbound IP address
Connections that Clikma initiates to a company's systems come from a fixed IP address, which the company's IT team can add to its allowlist.
Hosting, high availability and backups
The production platform runs in Google Cloud, in the Santiago de Chile region (southamerica-west1), on managed Kubernetes.
Multiple replicas per service
Each production service runs with more than one replica, and an availability policy keeps at least one active during node maintenance.
Regional database with a standby replica
The production database keeps a standby replica in another zone of the region, which takes over if the primary fails.
High-availability sessions
User sessions live in an in-memory store with a replica in another zone.
Daily backup and point-in-time recovery
The database is backed up every day and can be restored to any moment in the last 14 days.
Daily configuration backup
The cluster configuration is backed up every day, and a documented, scripted procedure makes it possible to bring up the entire platform on a new cluster.
Scheduled maintenance
Maintenance windows are on Saturdays and Sundays from 04:00 to 10:00 UTC.
Monitoring and operations
The Clikma team monitors each service's availability, resource usage and the database, and reviews the logs when something is not working as it should.
Availability monitoring
Each public service is monitored continuously, with a status page for the operations team.
Metrics and alerts
Alerts on the database and on the services' resource usage.
Centralized logs
Logs from all services are gathered in one place to investigate an incident.
Tracking of every email sent
Delivery, open, click, bounce and complaint for every email.
Adoption and actual platform usage
Active users, concurrent sessions, peak hours and devices.
How changes are developed and released
Every change follows the same path before it reaches client companies, and the infrastructure is declared in versioned repositories.
Separate environments
Development, demo, user acceptance testing and production, each with its own data.
Review and approval
Every change comes in through a merge request that is reviewed and approved before it is merged into the production branch.
Static code analysis
Each service is analyzed before it is merged, to detect bugs and vulnerabilities.
Accessibility checks on every update
Automated rules check the accessibility of every change; a change that does not pass is not merged.
Automated deployment
Each service is built and released through its own pipeline, always the same way.
Scale in production
These figures come from Clikma's production databases.
20,540 people in one day
On September 2, 2026, with more than 10,400 people in a single hour.
More than 46,000 people every month
Distinct people who log in to Clikma each month.
More than 400,000 registered users
At organizations in mining, retail, transport, healthcare and other industries.
Frequently asked questions
Where is the Clikma platform hosted?
In Google Cloud, in the Santiago de Chile region (southamerica-west1), on managed Kubernetes, with the production database in that same region.
Is Clikma a multi-tenant platform?
Yes. Each company has its own subdomain, branding, configuration and logically separated data on the same platform, without separate installations, and all of them receive the same updates.
What technologies is Clikma built with?
Services in Java with Spring Boot, an Angular web application, its own content players, PostgreSQL databases, Redis for sessions and a local language model for the AI assistant, all on Kubernetes.
Does Clikma have an API?
Yes. Each company uses the API with its own credentials, an IP allowlist and privileges defined for it, and Command Center indicators can also be queried through the API.
What happens if a data center zone fails?
Services run with more than one replica, and the database keeps a standby replica in another zone of the region, which takes over if the primary fails.
How often is the data backed up?
The database is backed up every day and can be restored to any moment in the last 14 days. The cluster configuration is also backed up daily.
How is Clikma updated?
Every change goes through development, code review, static analysis and test environments before it reaches production, and it is released through automated pipelines. Scheduled maintenance takes place on Saturdays and Sundays from 04:00 to 10:00 UTC.
What IP address does Clikma use to connect to my systems?
A fixed outbound IP address, which the Clikma team provides to the company's IT team so they can add it to their allowlist.
Next steps: how to get started with Clikma
If you are evaluating a platform to train, assess and measure your people, this is the usual path.
Book a meeting
On the contact page you pick a time on the calendar, or you can write to contact@clikma.com or message us on WhatsApp at +56 9 7960 1634, Monday to Friday from 09:00 to 18:00.
See a demo with your use cases
We walk through the features you care about, with your processes and your numbers.
Define the scope and implementation
With a project manager, the scope, timelines and roles are set; the HR system is connected, single sign-on is configured and the training history is loaded.
Pilot run and ongoing support
Before opening it to the whole company, everything is checked and adjusted in a pilot run; after that, the Clikma team keeps supporting the operation.
Get to know Clikma with your own use cases
In a demo we walk through the features you care about, with your processes and your numbers.
Book a Demo