Aug 2022 – Feb 2024 · 1 yr 6 mos

• Built an MVP to facilitate in person meetup on mercari
• Revolutionised the way Microservice PRs are tested at Mercari with my initiative Simplify Microservice Testing. Average development time across Mercari reduced by 20%. Earlier, to test any microservice PR, we had to update environment variables of all other dependent microservices, which meant creating extra PRs for all of them. I instead built a dynamic routing system using client interceptors in Golang based on headers set via clients.
• Delivered two brand new shipping options which offered a super cheap of delivering Trading Cards and Books. I led this project and delivered it within a month.

Jan 2022 – Aug 2022 · 7 mos

• Built an MVP for furniture delivery at Mercari. I designed and built all the backend systems for this project. In addition, I was also the project lead. I handled communications with all involved stakeholders, built a roadmap and made sure we adhere to it at all times and made dynamic adjustments as per changing product priorities.
• Increased productivity of my team by serving as scrum master and making improvements every sprint.

Feb 2021 – Jan 2022 · 11 mos

• Delivered Mercari Local, a same day delivery product which is currently a hot topic in the c2c e-commerce industry
• I worked in a team with stakeholders in multiple timezones and delivered a product used by millions of users across the US in less than a year
• In addition, I also delivered a QR code substitute for shipping labels single handedly.

Oct 2019 – Feb 2021 · 1 yr 4 mos

• I am currently working as a backend engineer in the shipping team
• Some of the features I develop are in PHP and I build microservices in Golang
• Writing tests, playing around with data and handling deployments is also a part of my job

Jul 2019 – Aug 2019 · 1 mo · Bengaluru, Karnataka, India

• I worked as a full stack engineer using React for frontend, Django for backend and pytorch for ML related stuff
• I built an image tagging platform for our doctors to manually annotate cell images
• Using the images tagged in the above model, I trained a classification model to classify WBC types

Jul 2018 – May 2019 · 10 mos

• Exploring approaches for Machine Translation of low-resource languages

May 2018 – Jul 2018 · 2 mos · Tokyo, Japan

• The goal of my project was to build a platform for Similar Image Search suitable for mobile devices.
• We worked in a team of two out of which, my job was to extract feature vectors from the images and the other part was to do a fast search using a novel indexing technique.
• We successfully completed the project and at the end I also created a demo web application.
• The code was written using tensorflow for python and I used Flask to create the web application.

May 2017 – Jul 2017 · 2 mos · Urbana-Champaign Area

• Implemented a magnetometer based attack on a 3D printer in Matlab using Digital Signal Processing techniques. Also developed a random string approach as a to defend this attack.
• Developed and implemented an Image Processing based algorithm for segmentation which helped in automating the attack.
• Lot of experimentation and brainstorming was involved and hence it was fascinating. The way the project directly mapped physical concepts to theoretical cryptography was quite remarkable.

May 2016 – Jun 2016 · 1 mo · Kharagpur Area, India

• Worked as an intern for this National project intended to create a free digital library that can be accessed all over India.
• My job was to automatically classify a new book into its appropriate category based on its Metadata fields (Dewey Decimal Classification). I designed the algorithm for it and wrote the code in java using the mediaWiki API.

Mar 2015 – Jul 2017 · 2 yrs 4 mos · Kharagpur Area, India

• I work mainly in the field of path planning. I have implemented two different path planners, one of which was implemented using the MRPT library, and the other using the move_base framework of ROS.
• I have also designed the lane_navigation algorithm of our vehicle.
• Because of my dedicated efforts, and my knowledge of the entire data flow required for the working of the vehicle, I was selected to be a part of the 4 member team to represent our institute at the Intelligent Ground Vehicle Competition (IGVC) 2016, held at Oakland University, Michigan.