Get Started

Mandarin Platform Project Takeover and Recovery

Stralya was appointed to rescue an existing Mandarin training platform originally developed by a third-party vendor. Within weeks, we secured the codebase, re-architected critical cloud-native components and restored the reliability, performance and delivery roadmap of this high-stakes e-learning product.

Client

MandarinMaster

Location

France

Industry

EdTech · E-learning · Language Training

Engagement Model

Fixed-price project rescue with ongoing expert maintenance and evolution.

Scope of Work

Cloud-native audit and takeover, security hardening, refactoring of critical services, performance stabilisation, DevOps and release process structuring.

Technologies

Modern cloud-native web stack (container-based services, managed databases, CI/CD pipelines) aligned with AWS/Azure best practices.

About

The Mandarin Platform Project Takeover and Recovery was initiated when the client’s existing vendor could no longer deliver to the required standards of reliability and scalability. The platform, a cloud-hosted Mandarin language learning solution, had already attracted a fast-growing user base but was increasingly fragile.

Where digital experiences are expected to be seamless, secure and always-on, the client needed an expert partner capable of stepping in quickly, taking full ownership and restoring confidence in the product. Stralya was engaged as a boutique cloud-native web development partner with a clear mission: secure the platform, stabilise operations and structure the technical foundation for long-term growth.

Working under a fixed-price, outcome-driven model, we took over the existing codebase, infrastructure and delivery pipeline. We approached the engagement with a “project-first” mindset, focusing not only on the code but on the business-critical outcomes: availability, performance, data integrity and the ability to safely roll out new features without breaking the learning experience.

This project is a representative example of Stralya’s project rescue capability in the UAE: stepping into complex, partially delivered web platforms and turning them into reliable, cloud-native products aligned with international standards.

Problems

When Stralya was called in, the Mandarin training platform was already live but increasingly unstable. The client faced several critical issues that threatened both user trust and business continuity:

  • Lack of ownership and accountability – The initial vendor operated on a man-day, task-based model with no real responsibility for outcomes. Releases were unpredictable, and issues were often left unresolved.
  • Unstable core architecture – Key services such as authentication, lesson progression tracking and content delivery had been implemented without clear boundaries or scalability in mind. Under peak usage, the platform experienced slowdowns and intermittent outages.
  • Security and data risks – Sensitive user data, including learning progress and personal information, was not sufficiently protected. Access controls were inconsistent, and logging was incomplete, making incident analysis difficult.
  • Fragile deployment process – Releases were manual and error-prone. A single deployment could trigger regressions in unrelated parts of the application, and rollbacks were complex and time-consuming.
  • Incomplete documentation – The previous team had provided little to no technical documentation. Knowledge was fragmented, and there was no reliable map of dependencies between services, databases and external providers.

The combination of these factors created a high-risk environment: any new feature or configuration change could cause a critical incident in production, and the internal team lacked the senior technical capacity to safely manage the situation.

Solutions

Stralya approached the Mandarin Platform Project Takeover and Recovery with a structured, cloud-native and outcome-focused methodology tailored to high-stakes digital products in Dubai and the wider GCC.

1. Rapid audit and technical takeover
We began with a time-boxed, in-depth audit of the existing platform:

  • Mapped the full architecture, from front-end components to back-end services and cloud infrastructure.
  • Assessed code quality, security posture, performance bottlenecks and operational risks.
  • Identified the truly critical paths: user authentication, course access, payment integration and learning progress tracking.

This allowed us to quickly define a stabilisation roadmap and assume full technical ownership without disrupting day-to-day operations.

2. Securing critical components and data flows
Given the sensitivity of user data and the importance of business continuity, we prioritised security and reliability:

  • Hardened authentication and authorisation flows, aligning them with modern security practices.
  • Reviewed and tightened access to databases, storage and admin interfaces.
  • Implemented structured logging and monitoring to detect anomalies early and trace incidents accurately.

These measures significantly reduced the risk of data leaks, unauthorised access and silent failures.

3. Refactoring and isolating core services
Instead of a risky full rewrite, we applied targeted refactoring to the most fragile components:

  • Isolated critical services (such as progress tracking and content delivery) behind well-defined APIs.
  • Reduced hidden coupling between modules to prevent cascading failures.
  • Improved error handling and fallbacks so that non-critical issues would not compromise the entire learning experience.

This approach delivered fast, visible reliability gains while keeping costs and timelines under control.

4. Structuring a robust cloud-native delivery pipeline
To eliminate deployment-related incidents, we restructured the delivery process:

  • Introduced a CI/CD pipeline with automated tests on key user journeys.
  • Standardised build, test and release steps for all environments.
  • Implemented safer deployment strategies, enabling quick rollbacks when needed.

The result was a predictable, repeatable release process aligned with international cloud-native standards.

5. Performance stabilisation and observability
To support a growing user base across time zones, we focused on performance and observability:

  • Optimised database queries and caching strategies for high-traffic features.
  • Added metrics and dashboards to monitor response times, error rates and resource usage.
  • Set up alerts to proactively detect and address issues before they impacted learners.

This observability layer gave both Stralya and the client clear, real-time visibility into platform health.

6. Long-term partnership and evolution plan
Beyond the initial recovery, we worked with the client to define a sustainable roadmap:

  • Prioritised future features based on impact and technical feasibility.
  • Recommended governance practices for change management and quality assurance.
  • Established a long-term maintenance and evolution framework, ensuring the platform remains secure, scalable and aligned with business goals.

Throughout the engagement, Stralya acted as a strategic partner rather than a simple vendor, taking full responsibility for outcomes and ensuring that the Mandarin platform could confidently serve its learners in Dubai and worldwide.

Une entreprise avec un vrai esprit entrepreneurial, rapide, professionnelle et surtout réactive.

Phillipe Laurent

CEO, MandarinMaster

Let’s Build Something Great

Tell us about your project, your goals, and your vision. We’ll take care of the tech, performance, and delivery.

Start Your Project