Feb 2022 – Dec 2022 · 10 mos · Remote

• ◉ As a part of the `Assist` team, responsible for developing/managing IUR(IVR + Chat) platform(s) to be used by `customer support agents to handle real-time, inbound & outbound interactions(Call/Chats/Email/SM) from customers.
• ◉ Memory/CPU optimization - Reduced the total number of Java threads from 3K to 500, and open the number of file descriptors from 17K to 4K
• ◉ Migrating IVR infra from Twilio to internal telephony systems to reduce the cost.

Oct 2021 – Jan 2022 · 3 mos · Remote

• ◉ Developed `Quality & Audit Tool` which helps in auditing the interactions(Calls/Chat) b/w Udaan agents and the customer and each scorecard is generated for each interaction.
• ◉ Contributing to the `Decision Tree(DT)`. DT helps Udaan agents to get to the solution of the query based on the inputs provided by the customer.
• ◉ Added VOC(Voice of the customer) based filter in DT, which enables agents to ask the question to customers to find out the appropriate solution.
• ◉ Developed an automatic dialer to place outreach calls, which will be used by relationship managers/Sales guys to reach out to the org(customers). Org/Customer details will be assigned to online OUTREACH agents and then place a call.
• ◉ Developed functionality to move(Cut/Copy) a DT node anywhere within the tree as well as across multiple trees.
• ◉ Upgrading Cosmos DB
• Tech stack: Kotlin, Redis, CosmosDB, React.js, Twilio

Apr 2020 – Oct 2021 · 1 yr 6 mos · Bengaluru, Karnataka, India · Remote

• ◉ Kafka infrastructure: Deploying Kafka on all environments and provided producer/consumer library on top of Kafka client(Java).
• ◉ Cassandra Migration: Migrating from Cassandra 1.2 to Cassandra 3.11
• ◉ Developed in-house Cassandra schema synchronization/versioning utility.
• ◉ Worked on a framework to migrate PII to encrypted format in Cassandra.
• ◉ Improved the database triggers so that if any trigger fails, it’ll be retried later(via Kafka consumer) to avoid data inconsistencies. In case of Irrecoverable failures, an alert will be raised based on the severity.
• ◉ Developed a framework to handle data inconsistency.
• ◉ Written ansible roles for Kafka, Zookeeper, Cassandra.
• ◉ Introducing Vault for storing secrets/passwords, internal teams will act as its consumers.
• ◉ Involved in encryption key rotation & re-encrypting data across the DB.
• Tech stack: Java8, Python3, Cassandra(1.x & 3.11.x), Redis, Kafka, Docker, Ansible

Feb 2019 – Apr 2020 · 1 yr 2 mos · Bengaluru, Karnataka, India · On-site

• Dataweave platform -
• ◉ Pricing intelligence & Brand analytics.
• ◉ Worked on diagnostics APIs to help the monitoring team.
• ◉ Worked on “Discoverability” which comes under ‘Brand Analytics.’ Discoverability comprises of availability/stock analysis, keyword/visibility analysis & content analysis of the data.
• 1. Availability/stock analysis - Computes the availability score of each product on the basis of its stock status across all the retailers in all the stores.
• 2. Keyword analysis - Computes visibility score of each product for a given keyword which gives an idea that how discoverable a product is for that keyword.
• 3. Content analysis - Computes content quality score of each product which depends on the factors like if a product has the desired title, description? are the product images of good quality? Comparisons are done with the reference catalog provided by the client(brand)
• Tech stack - Python 3.6, Tornado, Redis, MySQL, ELK, RabbitMQ, Kafka

Dec 2017 – Jan 2019 · 1 yr 1 mo · New Delhi, Delhi, India · On-site

• ◉ Involved in developing a product to perform cleaning services with ease by capturing various details e.g. cleaner's info, cleaning time, working time, presence time, etc.
• ◉ Developed a few microservices: Events engine, Subscriber (Explained below in separate section)
• ◉ Developed a notification engine using Python, Sanic(Async/Co-routine framework), MySQL, RabbitMQ. So, users can subscribe to the event(s), and when those event(s) occur, actions/notifications(Webhook, Email, SMS) will be triggered in the background to notify all the subscribers.
• ◉ Generic utility/framework to convert JSON objects to SQLAlchemy queryset.
• Link: https://github.com/suyash248/sqlalchemy-json-querybuilder
• ◉ Generic utility to filter, sort and export the data from different data sources(REST endpoints, RDBMS) in different formats(PDF, Excel, HTML, CSV). HTML & CSV files are exported via HTTP streaming.
• ◉ Developed the authentication & authorization layer to authenticate the user from different sources, and then checks for authorization on the basis of current role/permissions via ACL(Access Control List).
• Products:
• ◉ Smartview (Time & Task): Developed a platform for clients/customers to perform/monitor cleaning and maintenance tasks at various sites/buildings and later enables them to do the analytics on this data.
• Tech stack: Python 3.6, Flask-Restful, Redis, MySQL, ELK, Gunicorn
• Microservices:
• ◉ Events engine: Enables subscribers to subscribe to the event(s), and later when an event occurs, all the subscribers will be notified via webhook or email.
• Tech stack: Python 3.6, Sanic, Celery, Redis, Gunicorn, aiomysql
• ◉ Subscriber: Enables RFID readers to send data in the textual format, over TCP connection, passed thru a middleware that converts textual data to objects/JSON later performs the validation on that data. Then this data gets processed and stored in Cassandra & MySQL in a suitable format.
• Tech stack: Python 3.6, aiohttp, Celery, Cassandra, MySQL, Gunicorn, aiomysql

Oct 2015 – Nov 2017 · 2 yrs 1 mo · New Delhi Area, India · On-site

• Customer engagement SaaS-platform.
• Tech stack: Java 1.8, Play framework 2.5, Akka, Scala, Redis, Cassandra 3.x, PostgreSQL., Python, Flask, Celery, Gunicorn, Node.js, Socket.io
• Contributions:
• ◉ Cassandra Infrastructure: Scaled the services 4x by replacing PostgreSQL with Cassandra & migrating data from PostgreSQL. Developed in-house lightweight ORM for Cassandra.
• ◉ Upgrading to Play framework 2.5, plugged-in Authentication & Authorization(RBAC & ACL) layer.
• ◉ Redis Infrastructure: Bringing in In-memory DB(Redis) to our tech stack. As a result, the dashboard loading time decreased to ~3x
• ◉ Development of `Actionable insights`, which enables clients to create automated rules for targeting their users through multiple channels(Browser push notification, Onsite push notification, Webhook, Email, SMS, Exit intent).
• ◉ Developed `Exit intent`: Keeps track of user activity across sessions and shows pop-ups(configurable) based on the rules(related to user’s location, OS, browser, activities, time spent) created by the client.
• ◉ Built payment module supporting customized plans and coupons management.
• ◉ Beam: “Microservice”, For sending browser push notifications on Chrome, Safari, Firefox. Tech stack includes Python 2.7, Flask, Celery, Redis, Gunicorn
• ◉ Ray: “Microservice”, provides a fully-duplex(2-way) communication platform via WebSockets. Tech stack includes Node.js, Socket.io, Redis

Jun 2014 – Sep 2015 · 1 yr 3 mos · Dharamshala, Himachal Pradesh · On-site

• Role: Software Developer(Java/JEE)
• Highlights:
• ◉ Developed LMS using Java/JEE, Spring, Struts2, JPA.
• ◉ Developed a real-time dashboard using Flot.js and RESTful APIs.
• ◉ Worked on User profile & leads management module.
• ◉ Authentication & authorization using Spring security.
• Products:
• ◉ Examjockey
• ◉ ABCTE