Building A2A (Agent-to-Agent) AI platform that enable autonomous AI agents to collaborate and execute complex workflows.

Bar Raiser @ Uber with over 250 interviews in the last couple of years to boost engineering excellence across the site.

I was responsible for building, scaling, and architecture of our customer platform, the software stack used by Uber’s customers, employees, and partners that helps us effectively address these defects both proactively or in response to an issue raised by a customer ultimately resulting in customer delight!

In essence, CO owns the entire customer support infrastructure for the global transportation marketplace that spans a wide range of cities — varying from 100,000 to 10 million people in size — with unique regulatory, technical, and business requirements.

I was working in the Customer Obsession team.

We are dedicated to improving the Uber experience by developing in-app support, self-service flows, and other technologies that make contact resolution faster and more seamless.

I worked in the Core Platform Development team.

Synaptic is a Vy Capital backed early stage Fintech startup, which is an alternative data platform that helps financial firms and investors get actionable insights from vast amounts of data.

Technologies: Python (Django/Flask), Ruby (Rails), MySQL, Amazon Redshift, PostgreSQL, Cassandra, MongoDB, Redis, ElasticSearch and AWS.

I worked in Innovation Team (R&D).

Being in a product based startup, I got to work on the bleeding edge of the Shuttl spectrum. It was a full stack developer role which requires developing and designing each and every bit including hardware, front end, back end, design, databases, project management.

• Redesigned the Lead management system from grounds up for better efficiency, which helped in increasing the lead conversion rate by 58%.

• I helped in designing the network optimization system from grounds up so as to be able to schedule the buses efficiently on the route map to minimize the wait time delta between trips and maximize the utilization of the whole network. It helped in increasing revenue by 1.4x without adding any additional resources (buses) on road.

• Successfully implemented core features of Project Shuttl S.A.F.E. as an individual contributor: Auto onboarding authentication via face detection and recognition, ShuttlTV, Always on Alcohol detector with ignition cut off and all of this powered by a solo Raspberry Pi 3. This project won FICCI's "Road Safety Product & Solution" 2017 special jury award.

Technologies used: OpenCV, DLIB, Python, Flask, Raspberry Pi, Arduino, Javascript, Data analysis, Data mapping and visualisation