Saket Saumya

• Data Structures (A-)
• Design and Analysis of Algorithms (A-)
• Cloud Computing (A)
• Operating System (A-)
• Database Management Systems (A)
• Information Theory And Coding (A-)
• Bachelors Project (A-)
• Masters Project (A-)

• Eliminated 4xx P0 production outages in automated chat system by building an Event Processor in Golang using RabbitMQ as distributed messaging queue to operate within specified rate limit.
• Reduced 25 % of Customer Delight Team chat volume by building an Infra for managing the premium plan susbscriptions and introduce features Coach Change and Plan Pause/Resume right in the app.
• Delivered systems for coach assignments and migrations in bulk. Reduced TAT of the process by 87 %. Enhanced dipping Renewal rates and NPS of coaches.
• Improved tail-latency P98-P99 of some high throughput legacy MySQL DB transactions using indexing, better query planning and batching to avoid multiple network RTTs.
• Authored around 12 TRDs, ensured 90%+ code coverage in modules and used appropriate design principles.
• Python (Django) · Golang · AWS · MongoDB · GraphQL · MySQL · Redis · RabbitMQ

Received Full-time Offer (PPO) on the basis of performance.

• Led design and development of a dynamic feature flag for Payments team in Java and Spring. This flag removed mandatory login and reduced payment completion time by over 60%.
• Integrated migrated IOT Devices APIs (ex-Soundbox, Smart POS etc) with merchants service as a part of Payment Gateway 2.0 migration.
• Implemented event driven Kafka based APIs dealing with 20M+ online and offline merchants communication channels, URL configurations and notification settings. This reduced tech debt by 4%.
• Tech Stack: Java · Spring Boot · Apache Kafka · MySQL · Elasticsearch · Redis

• Decoupled services using Pub/Sub pattern with Python, Redis and Kinesis. The new system was able to asynchronously valuate 3M+ Real Estate properties in bulk.
• Improved CI-CD pipelines and AWS Infrastructure. This reduced developers on-boarding time from a day to 30 mins and new features release time by 60% using 1-click deployment functionality.
• Tech Stack: Python (Django) · Redis · Celery · CI/CD Pipelines · AWS (ECS and Others)

Received offer to work for Bihar Government - Revenue And Land Reforms by IT Head of Department.

• Scraped LRMS website involving asynchronous loading, authentication and paginated response.
• Architectured Flutter based Android App for making CRUD operations easier with scraped data stored in Realtime DB for the better management of Land Records cutting paperwork by 80%.
• Tech Stack: Python · Flutter · Cloud Firestore · Web Scraping

Authoring a book on AsyncIO to help spread awareness in Python community about using asynchronicity in backend web applications.

MacOSX Menu Bar App (200+ installs) removing native OS restrictions on battery information.

Flutter Android App (approx. 100 installs) to register, match students & teachers based on requirements. Deployed on Play Store as well.

Web App that register students, takes attendance through webcam and updates in Excel asynchronously.

A project submitted in partial fulfillment of the requirements for the the award of the degree of Bachelor of Technology under Prof. Shashikala Tapaswi and Prof. Manisha Pattanaik

A project submitted in partial fulfillment of the requirements for the the award of the degree of Master of Technology under Prof. Shashikala Tapaswi.