Jun 2022 – Sep 2022 · 3 mos · Menlo Park, California, United States

• Facebook group recommendation ranking team
• 1) Developed an end-to-end machine learning pipeline to improve the ranking of Facebook related group recommendations.
• 2) Designed several algorithms to increase diversity in recommendations.

Sep 2021 – Jun 2022 · 9 mos · San Diego, California, United States

• Graduate Student Researcher working on applying deep learning methods to cancer genomics. Focused on methods to apply for interpretability in deep learning models.

Mar 2021 – Apr 2021 · 1 mo

• Worked on an Omdena AI challenge in collaboration with NeedEnergy(an energy tech startup) and 40 global collaborators to improve access to clean energy in sub-saharan Africa through predictive modelling.

Jul 2019 – Aug 2021 · 2 yrs 1 mo · Pune Area, India

• ° Used Python and OpenCV to process images of text and extract required fields and text from the images.
• ° Developed a portal for global and industry-leading client account onboarding and lifecycle management solution to HSBC. It allows for fast, secure and efficient data exchange to enable a more efficient and streamlined client lifecycle experience with the bank. Used PostgreSQL, Java and React.
• ° Delivered an end-to-end customer journey for an urgent Covid-19 requirement- US Paycheck Protection programme within 8 weeks amidst changing US govt guidance. After 1600 production releases by my team, a digital channel was provided to facilitate the end-to-end loan forgiveness process.
• ° Participated in multiple hackathons including the Grads Hackathon, collaborating with Technology Grads from HSBC Technology Centers across the globe. Won the Microsoft partner prize in the Hackathon for the solution- interactive voice assistant for the best execution of human-computer interaction via Voice (using Microsoft Azure Cloud Services) designed to improve banking experience.

May 2019 – Jun 2019 · 1 mo · Pune Area, India

• Designed a system to allow product and services teams to create and visualize their Value Stream Mapping. Used REST API's to extract meaningful data from various tools and used Python to aggregate and analyse the data. React was used for visualizing the CI/CD pipeline. The configurable automated system is meant to provide a generic and reusable approach that caters to different pipelines and allows the analysis of value that each process within the workflow brings.

Jan 2019 – Feb 2019 · 1 mo · Kolkata Area, India

• Worked on building an analytics dashboard using Microsoft power BI after performing sentiment analysis of Facebook data using the Text Analytics API of Microsoft Azure Cognitive Services.

May 2018 – Jul 2018 · 2 mos · Montreal, Canada Area

• As an intern under the Mitacs Globalink Internship Programme, I interned at ETS, Montreal, Canada(Département de génie logiciel et TI) for a period of 11 weeks under Professor Roberto Herrejon Lopez. I worked on a research paper which focuses on the customization of technologies for autistic children. I worked on a systematic mapping study of customization capabilities of assistive technologies. Also, studied two wearable technologies(HexoSkin and Empatica E4 watch) for capturing various bio-signals and storing them for further analysis. Was able to procure real-time data from the Empatica E4 watch and documented ways to analyze the data. The research was published in an international journal.

May 2017 – Jun 2017 · 1 mo · Mumbai Area, India

• Eklavya Summer Internship Programme
• Under the guidance of Professor D. B. Phatak, Computer Science Department, IIT Bombay
• Built a flexible web application for an enhanced group learning environment. This online platform is equipped with a range of group learning tools such as assignments, notes, quiz, group discussion forum, video conference and chat etc which serve for an enhanced virtual learning and sharing experience. Each tool can prove an in depth analysis of the participation and performance of the group members. The USP of this application is the flexibility in which collaborative groups can be created along with tool selection and various levels of access rights which can be allotted to each profile for every tool.

Jan 2017 – Jun 2017 · 5 mos

• Developed a social task manager Android app using Java and XML as part of the Aakar i3 internship programme. The app manages groups of people, keeps track of task status, send emails and notifications.

Aug 2016 – Jan 2016 · 5 mos · Bhubaneswar

• Created a cognitive application using IBM Bluemix.The application had a “Natural Language Classifier” service which was trained using a sample set of texts from tweets indicating complaints, queries or feedbacks. The application was based on Node-red which read twitter mentions of a particular handle, used the NLC Service to identify the type of tweet and based on the type of tweet, automatically replied to that tweet with an appropriate response.

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

• Developed an Android app using Java and XML which calculates the employee productivity score for an employee based on the number of hours worked per week. Authentication with Google sign in was enabled using Firebase.