Jan 2025 – Apr 2025 · 3 mos · Bengaluru, Karnataka, India · On-site

• Build Rest api using spring boot and java to make trade finance digital
• Integrate external banking api from SBI and HSBC bank with in micro-service architecture to enable secure and seamless finance transaction.
• Built multiple functional nodes for a Finance AI Agent using LangGraph, enabling intelligent task execution such as text extraction, API mapping, and microservice orchestration.

Nov 2023 – Jan 2025 · 1 yr 2 mos · Bengaluru, Karnataka, India · On-site

• Full-Stack Data Science Website: Utilized React.js, Redux, Material-UI, Express.js, Node.js, and MongoDB to build a comprehensive data science website. Designed and implemented seamless API integrations for optimized frontend-backend communication, enhancing user experience.
• Automated Data Aggregation Pipeline: Developed a Python script using the PyArrow library to efficiently fetch data from HDFS, process it with pandas, and implemented logic to ensure that only unread files were processed. The script performed data aggregation and used batch statements along with parallel processing to insert the aggregated results into Cassandra DB, resulting in a 7x performance boost. Additionally, integrated logging features to capture info logs, error logs, and execution time logs. This solution was successfully deployed in a production environment.
• Data Processing Scripts for JVIZ Preprod: Developed and automated scripts to fetch, process, and store data in MongoDB. Rewrote codebase in PySpark for efficient handling and parallel processing of big data.
• Enhanced Data Management and Performance: Applied advanced techniques in Python, Pandas, NumPy, Cassandra, and PySpark to streamline data processing workflows. Optimized data storage and retrieval operations, significantly improving system performance and scalability.
• Automated Disaggregation Pipeline: Utilized the PyArrow and Pandas library to fetch and process data from HDFS. Developed a disaggregation pipeline using a seq2point machine learning model to predict the consumption of various household equipment based on hourly meter data. The results were efficiently pushed into Cassandra DB
• Automated Email Sending System: Automated the process of fetching disaggregation results from Cassandra DB, calculating total consumption for the month and the previous day, generates detailed visual plots of appliance-level consumption, and sending them through email
• Build Restful api using fastapi which upload data on server

Jul 2023 – Nov 2023 · 4 mos · Bengaluru, Karnataka, India · On-site

Oct 2021 – Dec 2021 · 2 mos · Japan

• Improved and debugged the current front-end of products and used ReactJS -
• Redux and CSS for a seamless interface.
• Leveraged the capabilities of the Sharp library to develop a robust functionality,
• allowing seamless extraction of images from Firebase storage, followed by
• efficient resizing and reuploading, optimizing the overall performance.

May 2021 – Jul 2021 · 2 mos · Jodhpur, Rajasthan, India

• Learned about an inductive dynamic graph analysis technique that can be
• potentially parallelized.
• Code Implementation of the technique based on graph neural network, feature
• extraction, Page-Rank and graphsage.

Apr 2020 – Mar 2021 · 11 mos · Jodhpur, Rajasthan, India

• Literature reviews different suspension systems .
• Set and calculate the parameter of suspension system’s component
• Create a CAD modeling for knuckle,hub, arm
• Analysis their life and sn curve on ansys software
• Software : solidwork, ansys,lotus