SUMEDH SINGH

Led architectural redesigns and implemented critical backend systems, optimizing email delivery and enhancing ancillary purchase processes for Airasia.

• Architected and deployed a high-level system, replacing the legacy Navitaire-to-email flow with an NMQP subscriber-publisher pipeline, enabling indirect PNR ingestion into Google Pub/Sub and streamlining email delivery via Communication Service.
• Developed an automated promo code application system for ancillary services (baggage, seats, meals, insurance), achieving real-time validation, maintaining utilization below 98%, and driving a 10% increase in ancillary purchases.
• Spearheaded migration from the legacy confirmation flow to the Zodiac-based Unified Confirmation Page, implementing key features like flight information display and check-in integration, enhancing user experience.
• Integrated JWT-based authentication across all Navitaire 4.5x calls by standardizing API clients with a custom header mechanism, ensuring secure and seamless token propagation.

Contributed as a Full Stack Engineer to the pre-booking experience, optimizing service communication and application performance with modern technologies.

• Migrated inter-service communication from REST to gRPC for critical microservices, significantly improving throughput and reducing payload size, resulting in lower latency and faster response times.
• Orchestrated the upgrade of Spring Boot from 2.x to 3 and Java from 11 to 17, leading to a notable reduction in application startup time and optimized resource utilization.
• Engineered multithreaded parallel fetching for ancillary services (insurance, meals, and baggage), replacing sequential API calls to reduce end-to-end response time and improve overall throughput for an enhanced customer pre-booking experience.
• Developed full-stack features for the pre-booking experience utilizing React, gRPC, and Java, enhancing user interface and backend functionality.

Developed an AI-powered solution for abstractive summarization and information extraction from complex insurance reports for clients.

• Built an LSTM model for abstractive summarization of insurance reports, simplifying complex data and improving client comprehension.
• Utilized Azure's Linguistic Service to implement Named Entity Recognition, achieving accurate and efficient information extraction from diverse data sources.
• Processed and structured complex data by filtering invalid responses, extracting JSON data, and organizing it efficiently using Pandas and regex.

Developed a real-time ride-hailing platform.

Developed an application for managing flight bookings, ticketing, and customer information.

Recognized for outstanding contributions and measurable impact within the organization.

Secured 1st position by building an online platform for shopkeepers to sell products.

Achieved 2nd rank in a competitive programming contest.