Assessment TeamMy team owns the platform that teachers use to create quizzes on Quizizz. Teachers use the quiz creation platform to create millions of quizzes and tens of millions of questions daily.In the past 1 year, we have heavily focused on reducing the time teachers spend creating questions/quizzes by creating an AI-first content creation and recommendation experience.I manage a team of 8 engineers across Backend, Frontend, and Machine Learning.Created the 0-1 journey for Quizizz's AI with 2 engineers before leading the entire quiz creation platform team.

Platform Team- Experimentation Platform: Introduced feature flagging in the org. This helped increase the speed of execution, the number of experiments that can be conducted and the time it took to conclude experiments across the company. Created an experimentation service that made it easier for all the services and clients to work with feature flags.- Data Ingestion Platform: Created a CDC-based data ingestion platform to improve the earlier data ingestion pipeline which relied on events manually triggered through code. There were many issues in the older pipeline leading to data loss, inconsistencies, etc. The new platform was designed to reduce any dependency on the developers and also provided certain guarantees to the data analysts making the data reliable.Monetization Team- Organizations in Quizizz for Work: Introduced the concept of organizations to Quizizz for Work thereby allowing users to add their colleagues on a single Stripe subscription. Org Admins can add/remove members and get charged on a prorated basis when buying new seats. Earlier, users had to purchase individual subscriptions with restricted collaboration on the content. The org plan enabled collaboration within orgs based on the user's requirements.- Game/Activity Scheduling: Designed the capability to allow teachers to schedule activities for their students. This helped the teachers save time during the classroom for live activities and also helped them schedule homework for different classes without having to do it after every class.

- Built a highly scalable SaaS product for conducting screening tests and coding contests similar to Hackerrank. Companies can create coding questions and tests and invite candidates to take them. The contest platform ran seamlessly for 1000+ concurrent participants while running memory and time-intensive code in our sandboxed online compiler. It can handle a much higher load during bigger hiring contests.- Built our online compiler (sandboxed environment to run code in supported languages). It worked as a plug-and-play product and could be configured to consume from different SQS topics and share the status and results/errors through webhooks. It is used by our online IDE, our coding practice platform, our self-paced courses, our online assessment and contest platform, our collab editor, etc. It is highly scalable and can scale horizontally based on the current queue lag.- Architected and built the backend of most of our products including our Coding Practice Platform, our SaaS product, and our Learning Management System (LMS) used by different bootcamps including Flipkart GWC, etc.- Managed and mentored 3 full-time employees and 12 interns (2-3 interns at any point in time).- Took care of marketing, sales, and operations scaling the company to 100k monthly unique visitors and 75L revenue.

Was part of the team that powers the order checkout experience of Flipkart.During my brief stint, built a feature that is significantly avoiding order cancellations and returns.The checkout system handled peak QPS of 16k during Big Billion Day (BBD) 2019. Did a small part in helping my team scale for BBD '19.Proactively helped Flipkart with their initiative #GirlsWannaCode to increase diversity in tech. Also, helped in talent branding by creating a job role and interview preparation document for SDE 1/2 roles at Flipkart.Had to leave within a year to work on my own startup.Exposure: Java, HBase, Kafka, Elasticsearch

Worked in the O365 Groups team.Played a major role in our team's initiative to start creating our processors using O365's microservice framework and also in migrating our existing web app to React.- Worked on creating a rich and interactive group card which is being used across most of the workloads (Outlook, Sharepoint, Yammer, etc) that use O365 Groups as an identity. The goal is to provide the user with the same experience across all workloads. The major feature that I worked on was developing a paginated member list with support for member management (promote/demote/remove) and server-side search to find and manage people in large groups.- Developed a microservice for creating and sending a welcome email whenever a user joins or is added to an O365 Group. The welcome email is supposed to increase the engagement of users towards various workloads which are based on O365 Groups.Exposure: C#, React, Redux, Exchange Architecture

Worked in the Search team.- Created an intuitive and simple to use query language for APIs which is used to make complex queries (expression tree with leaf filters and feature-rich multilevel sorts), complex aggregations (infinitely serial and parallel) and like Facebook’s GraphQL, obtain response based on specific attributes.- Had complete ownership over the Search API which serves as the backend for Tracxn's entire research platform, customer APIs, data exports, report generation, mobile apps, etc. The API is completely based on the abovementioned query language and is basically at the core of all the customer facing products. Exposure: Java, ElasticSearch, Kafka, MongoDB, Redis

Worked in the Search team.- Had complete ownership over the Search API which serves as the backend for Tracxn's entire research platform, customer APIs, data exports, report generation, mobile apps, etc. The API is basically at the core of all the customer facing products.- Actively worked on reducing developer efforts in building new features and/or modifying old ones. Proposed and created generic components for almost all the core search features after thorough analysis of possible use cases because of which new features get developed in minutes instead of days or weeks.Exposure: Java, ElasticSearch, MongoDB, Redis, Logstash, Kibana.

Started the programming club of my institute. Check out my quora answer to know how it has affected our college for good.https://www.quora.com/Are-there-any-good-online-forums-run-by-NIT-Durgapur-students-which-are-beneficial-for-students/answer/Gaurav-Chandak-1

Worked as an intern in the Personalization team- Developed processors for generating and serving personalized news feeds to users based on their usage and affinity towards certain newspapers, categories and topics.- Optimized the querying of news articles from ElasticSearch which helped increase the throughput of serving news feeds by 400%.Exposure: Java, ElasticSearch, MongoDB, Redis.

Worked as an intern in the NLP team- Developed a processor which classifies a corpus of news articles into topics and retrieves the top trending topics from it.- Earlier, users were sent hundreds of news articles for the companies/industries they were following of which most of the articles had the same news content.Exposure: Java, ElasticSearch, Tf-Idf, Latent Dirichlet Allocation, Stanford POS Tagger, Stanford NER