Jun 2024 – Sep 2024 · 3 mos · Carlsbad, California, United States · On-site

• Developed ML forecasting models on AWS to predict active network speed traffic patterns, optimizing bandwidth allocation.
• Designed ETL pipelines with SQL in Snowflake for models ranging from regressor to t-Transformers getting under 6 RMSE.

Dec 2023 – Apr 2024 · 4 mos · Irvine, California, United States · On-site

• Developing an advanced LLM-based chatbot to elevate student interactions, providing AI-powered responses to future inquiries.
• Building an AI based platform to automate the creation of sustainability dashboards based on AASHE data, with web integration via Streamlit to facilitate efficient querying and drive informed decision-making to enhance sustainability initiatives.

Sep 2023 – Jun 2025 · 1 yr 9 mos · Irvine, California, United States · On-site

• Information Retrieval under Professor Crista Lopez (Spring 2025)
• Information Systems under Professor Wu Fisher. (Winter 2025)
• Project Web Data and Applications under Professor J. Herold. (Fall 2024)
• Big Data Management Under Professor Vagelis Hristdis. (Spring 2024)
• Big Data Management under Professor Vagelis Hristdis. (Fall 2023)
• Holding Office Hours and providing support during hands-on lab sessions, troubleshooting issues,
• Performing review lectures on Spark, AWS, Kubernetes, Hadoop, SQL, and NoSQL concepts, helping students understand these key big data technologies.

Sep 2023 – Jun 2024 · 9 mos · United States

• Responsible for developing and implementing the image recognition algorithms for our waste image dataset, including image classification and image localization. The algorithms are used in our ZotBins mobile app as well as the physical Smart Bins hardware. Our daily work may include waste image collections, data labeling, data preprocessing, and algorithms development & implementation.

Jun 2021 – Aug 2023 · 2 yrs 2 mos · Mumbai, Maharashtra, India

• Built India’s most viewed OTT app, MX Player, completely from scratch with a customized workflow approach for various web platforms (mobile/desktop) and operating systems (LG/Samsung).
• Worked with the VideoJs library, Widevine license, and Dynamic Ads Insertion (DAI) to create a core HLS Streaming Player with inbuilt DRM protection and multi-layer ads.
• Created a User Authentication and subscription flow along with Rest API support on Java and payment service Just-pay for user management of 175 million people on the platform.
• Proposed and programmed an adaptive machine learning recommendation system that handles functionalities such as Continue Watching and Content Recommendation.
• Constructed internal games modules that handle (CI/CD) of Games correlating their versions with GitHub API
• Tech Stack - Frontend (Typescript, React), Backend(NextJs, Java), DevOps(Jenkins, NewRelic), ML

Jun 2021 – Jul 2021 · 1 mo · Bari, Apulia, Italy

• IoT based project Design and development of a smart environment with Professor Guiseppe Desolda

Aug 2020 – Apr 2021 · 8 mos · Kharagpur, West Bengal, India

• Under Dr. Pawan Goyal, Department of Computer Science and Engg, IIT Kharagpur
• Excel Depression: Multi-Modal fusion to predict depression on Social Media data self extracted and moderated over a period of the timeline to generate a big five personality characteristics score for different users state of the art transformers and LSTM architectures.

Apr 2020 – Jun 2020 · 2 mos · Hyderabad, Telangana, India

• Developed a back-end module in Java using Spring framework for improving errors clients joining the TCDA platform in Tier 3 and Batch fleets.
• Handled multiple databases and client sheets using Postgres to monitor and display discrepancies between container and Database API Calls.
• Designed a new algorithm to assimilate the exceeded Transactions per second ( TPS ) parameters and decreased the latency for the client portal.

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

• Visualized and preprocessed 100 advanced features from 3 lakh+ instances using pandas extracted from their cloud-based server
• Correcting the Label Imbalance problem: Implemented both label and label-powerset specific oversampling and undersampling algorithms that decreased value of the IRlbl and CVIR per label
• Used Cost sensitive learning methods improving Label Concurrence issues, achieving a 15% decrease in SCUMBLE value per instance
• Incorporated ML-ENN ,REMEDIAL and ML-SMOTE algorithms working on the multilabel implementation of Imbalanced-learn library
• Contrasted different Keras and Skmultilearn models improving precision and recall of neural networks by 35 % using Ensemble and Data Enrichment Techniques

Dec 2018 – Jan 2019 · 1 mo · Kharagpur Area, India

• Kharagpur Alumni Based Government funded Startup working towards Application of AI in Agriculture
• Deployed Open CV for removing Background Noise and improving edges with foreign matter using Image processing and Segmentation algorithms for 50 sheets per category for 4 categories of maize
• Tuned Mobilenet CNN model on Keras Framework using the concept of Transfer Learning for developing the workflow
• Improved Model recall and F-score from 60 % to 76 % using data augmentation and Blob detection techniques during preprocessing

Sep 2018 – Apr 2019 · 7 mos · Kharagpur Area, India

• Managed a Budget of INR. 30,000 across various technical events under Technology General Championship securing 1st position for the Hall.
• Member of Core Organising Team, for Rajendra Prasad Hall Cultural Night, Aahvan, which saw a participation of more than 200 boarders.

Aug 2018 – Apr 2019 · 8 mos · Kharagpur Area, India

• Technology Transfer Group (TTG) is a students' initiative under the aegis of Sponsored Research & Industrial Consultancy, IIT Kharagpur, which beliefs in the potential of the Institute as a premier Research & Development center. We act as a link between the industry and the academia to facilitate the transfer of industry-ready technologies and presenting IIT Kharagpur as a research consultant.
• Working as a bridge between professors and students of IIT Kharagpur for projects through UG Portal(Our Initiative to smooth-en this process )
• Successfully organized two workshops awaring the student community about Intellectual Property Rights(IPR).
• Lead a team of 10 members spearheading Publicity and Logistics for TEDx IIT Kharagpur 2019