Portfolio

Airhelp

AirHelp was founded in 2013 as a start-up and has grown to become the global leader in defending air passenger rights, helping over 13 million people underst...

#websites
Airhelp

AirHelp – Technology for the Leading Advocate of Passenger Rights

AirHelp was founded in 2013 as a start-up and has grown to become the global leader in defending air passenger rights, helping over 13 million people understand their rights and claim compensation for delayed, canceled, or overbooked flights. The company not only supports passengers through the compensation process but also fights for justice by representing them in legal disputes with airlines and lobbying for fair regulations at the government level. As a developer, I designed and implemented the AirHelp website, combining advanced functionality with technical optimization to meet the needs of a global user base. Below, I outline the project details, performance challenges, and solutions.

The Purpose of AirHelp and Its Audience

The AirHelp website was created for a company whose mission is to educate air passengers about their rights, simplify the compensation process, and provide legal support in 24 languages. The platform serves both individuals who are unaware of their rights or lack the time to take action themselves, as well as a broader community, thanks to collaboration with law firms in 30 countries and a team of 700 employees, including the world’s largest group of lawyers specializing in air law. My task was to create an intuitive, scalable website that supports these goals with modern technical infrastructure.

Technical Features of AirHelp

As part of the project, together with a 5-person team I led as Team Pilot, I implemented advanced technical solutions to meet AirHelp’s requirements:

  • Multilingual and Accessibility – Frontend architecture with Next.js and SSR framework, supporting 24 languages via i18n, compliant with WCAG 2.1 and optimized for mobile and desktop devices.
  • Compensation Process – Application form with dynamic flight data loading via GraphQL, integration with airline APIs, and transaction recording in a PostgreSQL database with AES-256 encryption.
  • Passenger Education – Informational section with legal articles loaded via REST API with caching in Redis and rendered in React for smooth performance.
  • Technical SEO – Optimization for keywords (e.g., “compensation for delayed flight”), dynamic XML sitemaps, and accelerated indexing via Google Indexing API.
  • Backups and High Availability – Automatic backups to Amazon S3 with regional replication, versioning, and Zstandard compression for operational continuity.
  • Performance – Server-side caching with Redis, Server Side Rendering, media optimization through Cloudflare with AVIF format, and support for HTTP/3 with QUIC.
  • User Dashboard – Application status tracking panel with real-time data via WebSocket, cached in Memcached for low latency.

Performance Challenges and Solutions

During the implementation of AirHelp, I encountered several key performance challenges that were solved as follows:

  • Database Load with Millions of Users – A large number of submissions caused PostgreSQL database overload. Solution: I implemented Redis for persistent query caching and database sharding with read replicas on Amazon RDS for load distribution.
  • Slow Loading of Application Form – Integration with airline APIs caused delays during peak traffic (e.g., after massive flight cancellations). Solution: I used RabbitMQ for asynchronous API query processing, with fallback to static data cached in Elasticsearch during timeouts.
  • High Multimedia Latency – Educational images and graphics slowed the website down on mobile devices in regions with weak connectivity. Solution: I used Fastly CDN with Brotli compression, WebP format, and lazy loading via Intersection Observer API, along with geo-optimization for faster distribution.
  • Delays in Real-time Dashboard – Status updates via WebSocket did not scale with 13 million users. Solution: I implemented Kafka for real-time data streaming with server throttling and AWS ALB load balancer for even traffic distribution.
  • Outdated Cache on Changes – New content and statuses did not refresh immediately due to aggressive caching. Solution: I implemented Varnish with custom VCL, purge on webhooks, and Edge Side Includes for dynamic sections, with additional cache busting through URL versioning.
  • High Resource Demand During Peak Hours – The site experienced overloads during mass flight disruptions. Solution: I configured auto-scaling on AWS EC2 with CloudWatch to monitor load, adding Cloudflare Rate Limiting for bot traffic protection.

Technologies Used

For building and maintaining AirHelp, I used the following technologies:

  • Yoast SEO – Metadata optimization, dynamic XML sitemaps, and search engine notifications for updates.
  • UpdraftPlus – Automatic backups to Amazon S3 with replication and AES-256 encryption.
  • Cloudflare – CDN with Argo Smart Routing, Brotli compression, and DDoS protection via rate limiting.
  • Redis – In-memory caching with sharding for sessions, forms, and dashboards.
  • Varnish – Server-side caching with custom VCL, grace mode, and ESI for dynamic blocks.
  • Lighthouse – Core Web Vitals audits integrated with CI/CD in Jenkins.
  • RabbitMQ – Task queuing such as API processing and email delivery, with retry and dead letter queue.
  • Elasticsearch – Flight and content search with fuzzy matching and data aggregation.
  • Fastly – CDN for parallel media distribution with geo-optimization.
  • Kafka – Real-time data streaming with partitioning for scalability.

Management and Technical Support

AirHelp is a project requiring continuous optimization and support. I regularly update the system and plugins, testing changes in a staging environment with backups on Amazon S3. Cloudflare, Redis, and Fastly ensure performance with global traffic, while Varnish, RabbitMQ, and Kafka stabilize dynamic processes. I monitor performance using Elasticsearch and CloudWatch, optimize SQL/NoSQL queries with complex indexes, and manage caching when changes occur. The platform can be expanded with integrations to ERP systems, an AI module for flight analysis, or a legal reports section, supporting AirHelp’s mission to fight for passenger rights.

Are you planning a website for your service company? Do you need a scalable platform with advanced technical support? Contact me to discuss the details and create a solution that meets your needs.