Jul 2023 – Dec 2023 · 5 mos · Bengaluru, Karnataka, India · On-site

• Contributing to the development of "𝐍𝐚𝐦𝐦𝐚 𝐘𝐚𝐭𝐫𝐢" based on ONDC Mobility Platform.
• ● Implemented edit pickup location feature.
• ● Implemented charging customer on ride cancellation feature.

Jan 2023 – Jul 2023 · 6 mos · Bengaluru, Karnataka, India · On-site

• Contributing to the development of "𝐍𝐚𝐦𝐦𝐚 𝐘𝐚𝐭𝐫𝐢" based on ONDC Mobility Platform.
• ● Implemented Alternate Number Feature for the driver app. Developed REST endpoints for validation of alternate number, verification of otp, resend otp and removal number.
• ● Implemented DTMF Feature for driver app to ease out calling process during the rides .
• ● Optimized driver pooling logic by reducing the Redis queries.
• ● Fixed customer side ride start and end locations by replacing older booking locations with actual pickup and drop locations when the customer changes his pickup or drop locations.
• ● Integrated Kapture , a Customer Support Platform to Namma Yatri for raising issues using tickets.
• ● Made feature for blocking fraud customers on the basis of rides.

Jun 2022 – Aug 2022 · 2 mos · Pune, Maharashtra, India

• ● Created frontend in JavaFx + integrated backend for internal tool which could convert around 𝟔𝟎% of the project from JAP based framework to Springboot. Adding frontend helped SDE’s to convert the project on a single click. On every task related to this the SDE was saving on an average 𝟒𝟓 𝐦𝐢𝐧𝐮𝐭𝐞𝐬.
• ● Integrated Git in the project with JGit using user credentials and Bitbucket location.This feature directly cloned the project to be converted from Bitbucket. Used multi-threading for processing backend tasks concurrently with UI frontend.

Jul 2021 – Oct 2021 · 3 mos · India

• ● Project: Semantic Segmentation of Urban features/buildings
• ● Tech Stack : Machine Learning , Deep Learning , Spatial Data
• ● Worked on the project Semantic Segmentation of Urban features/buildings. The project aims at fulfilling sustainable development goals and providing disaster management facilities as building type, size, shape play a significant part in the urban database model.
• ● Implemented FCN model for semantic segmentation of remote sensing images and extraction of buildings from them. For improving the results used CNN based Pretrained Model U-NET with different encoders like ResNet , InceptionV3 and VGG16 . Achieved an 𝐚𝐜𝐜𝐮𝐫𝐚𝐜𝐲 of 𝟗𝟔.𝟒% using InceptionV3 - UNET model.

Jun 2021 – Aug 2021 · 2 mos · India

• ● Project : Fine Grained Emotion Recognition for Counselling
• ● Tech Stack - Machine Learning ,Natural Language Processing, Image Processing
• ● Used Machine Learning algorithms and text-preprocessing techniques using Natural Language Processing to build it. Predicts the emotions of the person based on the text input by the person. Achieved accuracy of 74 %

Feb 2021 – May 2021 · 3 mos · India

• Contributed to Health Check Project 2021
• ● Worked on preparing Machine Learning Models
• ● Added Flask route for prediction of various diseases

Dec 2020 – Jan 2021 · 1 mo · India

• ● Worked majorly on the project "Statistics and Econometrics for Data Science" which seeks to simplify the understanding of some topics and provide code samples that can be used elsewhere for more robust answers in competitions/hackathons.
• ● Goals of Project - To make notebooks on the listed issues with features like introduction to the topic, mathematical proofs if required, code to implement that topic from scratch and any ready-made library code, if available.
• My Contributions
• ● I created two notebooks from scratch :-
• 1. Joint Probability Distribution- I started with the overview of the topic , added examples with graph visualizations and real life problems that can be solved using this concept.
• 2. Monte Carlo Methods for Inference- I worked on how estimations and predictions are made , hypothesis testing , Type I and Type II errors , using Monte Carlo simulations for real life applications like stock price prediction , power analysis and integration using Monte Carlo with graph visualizations, examples , explanation and solved a real life application for more exact and thorough overview of the topic.