Vikas Gupta

Designed and developed a scalable data streaming solution to stream different IoT device feeds in an asynchronous manner.

Distinguished features –

• Implemented a dynamic key management solution based on a symmetric encryption mechanism, that added security over TLS.
• Controlled access to resources, actions, and TTL (time-to-live) to devices using AWS Security Token Service.
• AWS Kinesis Data streams – Lambda integrations to publish the required notifications/alerts, while other consumers persist the information into MongoDB.
• Exposed server-to-server integration via set of microservices to serve the devices geo-locations and trip information
• Supports multiple IoT devices as producers, linked to multiple consumers.
• Utilizes the Kinesis retention period feature for the durability of the records.
• Migrated MongoDB standalone instance to clustered implementation to ensure high availability.

For cost-effectiveness and extensibility, the requirement was to come up with a scalable platform, which can provide WebRTC based video streaming solutions to multiple applications.

Solution - Designed and developed a multi-tenancy platform, using the Kinesis Video Streaming solution (WebRTC).

Distinguished Features –
• Communication dual channel – implemented PubNub messaging channel along with Kinesis Signaling Channel for real-time communications between publishers and consumers.
• Along with live streaming, for recorded media, the platform supports seamless integrations with AWS S3 and data streaming platform for management of meta-information.
• Implemented a dynamic key management solution based on symmetric encryption mechanism, that added security over TLS.
• Controlled access on resources, actions, and TTL (time-to-live) to devices using AWS Security Token Service.
• Saved more than 30% of the resources on the device side, by avoiding integrations with multiple solutions SDKs from different providers.

For British Airways, designed and implemented various system integrations to migrate the data flows from the mainframe system to the next-generation system.

Tasks Accomplished –
• Designed and implemented end to end integration patterns to standardize the data integration flows for different systems.
• Customized the Adaptris Interlok Framework by developing discrete components that helped expedite the system integrations by more than 50%.
• Mentored team to bring them quickly onboard on Adaptris Interlok Framework.
• Established the unit test process for the framework.
• Implemented the CI/CD process cutting down the phased release of the integrations.

A legacy modernization program aims to move the SITA Passenger Reservation System away from the mainframe USAS technology towards a next-generation solution.

Key Accomplishments –
• Actively involved in designing and developing passenger reservation and flight availability applications.
• Worked extensively on orchestrating a set of services over multiple applications to expedite go-to-market for a multi-channel sales platform.
• Introduced microservices architecture for a defined set of reservation services
• Utilized strength of Spring Security integrated with Oracle Identity & Access Management (IAM) to secure access to various services.

Designed and developed a fully distributed event-driven architecture to come up with a real-time surveillance system. Truly complemented by service-oriented architecture (SOA), the solution encapsulates the complex sub-systems. Middleware includes a multi-agent system distributed across different JVMs, still fully monitored, and serving various application components. Robustly designed to handle in-memory data structures efficiently.

Distinguished Features –
• Hardware integration cluster – A high performing apache storm/Netty architecture to handle more than 3 million hits per day without losing track of any of the transmitting devices. Proactive monitoring and alert mechanism to ensure that the stakeholders are informed
• Computation cluster – JADE-based dedicated computation architecture to process the analytical information and service the same to various components.
• Web cluster – Lightweight servers to serve web requests as well as mobile integrations. Spring-based architecture empowered with in-memory data structures to ensure a swift response with minimal delays.