Apr 2024 – Dec 2024 · 8 mos

• Working on making Gemini helpful to masses

May 2022 – May 2024 · 2 yrs

• Google assistant

Sep 2020 – May 2022 · 1 yr 8 mos · Bengaluru, Karnataka, India

• Enhancing customer and partner experience for 200M+ customers & 2.5M+ partners on the Ola Cabs app
• Laid the foundation for creating a seamless scalable partner onboarding experience with automation of document data capturing and validation
• Allocation Optimisation: Smarter cab-filtering for allocations
• Ride Use-case Identification to aid decision-making for the strategy team
• Traffic Lights Optimisation: Minimised wait-time for cars in the road network to mitigate congestion
• Semantic Search Development: Implemented a scalable food search autocomplete engine and semantic-based search engineer.
• LTV model with churn prediction: Built an improved LTV prediction model with added interpretability for people with high churn probability
• Others: In-house utility-packages (pyutilsds & rutilsds), Drop suggestion API, Conditional LTV model

May 2019 – Jul 2019 · 2 mos · Bengaluru Area, India

• Scalable Drop Suggestions Model: Impact: 0.4% increase in booking-completions
• Built a Gradient-Boosted Tree based drop-suggestion model adding better engineered time-based features. Improved on the previous in-production model by reducing error by 18.7%
• Spearheaded the development of one-touch booking system to facilitate hassle-free bookings

May 2018 – Jul 2018 · 2 mos · Pune Area, India

• Project:
• Spearheaded the development of life-science domain specific document summarization project.
• My major contributions:
• Prepared python scripts to parse MongoDB databases for analyzing, preprocessing and dumping PubMed documents for summarization
• Worked end-to-end with transformer model in TensorFlow for building a life-science domain specific abstractive summarization model
• Devised biomedical-tokenizer and integrated biomedical-encoder with the transformer model using the Unified Medical Language System ( UMLS )
• Implemented a graph-based sentence ranking system based on TextRank for extractive summarization of clinical trial (CT) documents
• Evaluated the extractive summarizer at an avg. ROUGE-L score of 31, efficient with summarizing a 30 pages CT document to a single page summary

Dec 2017 – Dec 2017 · 0 mo · New Delhi Area, India

• Worked on two major projects :
• Project on Prediction of Estimated Time of Arrival for a vehicle.
• Project on Route analysis and stoppage clustering of trucks across various routes in India.
• My contributions and work details :
• Examined 6 months of GPS-timestamp data for 800 trucks with MySQL and segregated useful information with feature engineering
• Constructed a neural network model for real-time predictions of estimated time of arrival (ETA) on subsequent toll booths using historical travel data
• Analysed more than 30 delivery routes and optimised road logistics for 250 trucks by stoppage clustering using density-based spatial clustering

Aug 2017 – Apr 2018 · 8 mos · IIT Kharagpur, Kharagpur, West Bengal, India

• Responsibilities I shouldered:
• Spearheaded conduction of Excavate, the Data Analytics Competition in 25th edition of the departmental fest, COMPOSIT 2018
• Shouldered management of around 200 participants during the fest and mentored a team of 20 core team members and 15 associate members

Dec 2016 – Dec 2016 · 0 mo · Kharagpur Area, India

• Successfully completed an IEEE certified workshop on Image Processing and implemented various Image Processing techniques and algorithms such as Hough Transform, Canny Edge Detection, Facial recognition etc. in OpenCv.
• Solved two problem statements
• 1. Tracking the number of goals scored in a Robo-soccer match.
• 2. Tracking the number of persons entering and exiting a room.

Dec 2015 – Dec 2017 · 2 yrs · IIT Kharagpur

• Responsibilities/Achievements:
• Worked on the design and analysis of Chassis of Eklavya 5.0 for IGVC 2016.
• Finalist team at Mahindra Rise Prize Driverless Car Challenge, built a self-driven car, Mahindra E2o.
• My Contributions:
• Spearheaded the design and building of chassis of our Autonomous Ground bot (Eklavya 5.0) for Intelligent Ground Vehicle Challenge (IGVC), 2016
• Performed stress-strain analysis of the bot chassis in ANSYS under various terrain and track running conditions and improved designs accordingly

Dec 2015 – Apr 2017 · 1 yr 4 mos · IIT Kharagpur

• My responsibilities and contributions:
• Successfully conducted Code-o-Soccer 2016, the only of its kind competition in India where soccer strategies are designed and implemented on robots
• Accountable for the administration of 21 playing teams in the competition held in Kshitij 2016 , Asia's largest Techno-Management fest
• Promoted the event in several colleges and garnered the participation of around 110 coding enthusiasts , the largest participation in history