Project: Oil & Gas Plant Digital Twin | Digitization Project
• Collaborated with a team of 6 members to implement an optimization algorithm for a virtual plant, considering variables like raw material prices, temperature, pressure, and duration to assess real-world feasibility.
• Actively contributed as a key member in the development of the Python backend logic for a digital twin of an oil & gas plant, facilitating effective simulation and optimization processes alongside team members.
• Designed the architecture for the digitization application of the Oil and Gas plant, ensuring seamless integration with existing systems while gathering input from various stakeholders.

Project: Real-Time Optimization | Machine Learning
• Worked collaboratively within a team of 3 members to enhance optimization algorithm performance with multiprocessing, achieving a remarkable reduction in runtime by 90%.
• Partnered with data analysts to develop Python code that fetches data from multiple sources, generating comprehensive reports in MS Excel for informed decision-making.

Project: Yield Optimizer Project | Differential Evolution Algorithm | Machine Learning
• Engaged in team efforts with 3 members to train machine learning models that predict propylene and ethylene yield from actual plant data, contributing to knowledge sharing and best practices within the group.
• Utilized these models in a DE algorithm to maximize the ethylene and propylene yield for an oil & gas plant, collaborating closely with process engineers for practical insights.
• Developed APIs using Flask/FastAPI and deployed them using Docker containers, ensuring smooth deployment workflows with the DevOps team.
• Played a key role in developing the backend for an API to send data for a dashboard from the database, working alongside frontend developers to ensure a cohesive user experience.

• Collaborated closely with a team of 10 members from the marketing team on customer segmentation using a clustering model, enabling targeted marketing campaigns tailored to specific customer needs.
• Analyzed marketing data with Python and shared valuable insights with the marketing team to enhance campaign performance measurement.
• Worked alongside stakeholders to build performance measurement dashboards for marketing campaigns, providing a comprehensive analysis of the target audience.
• Contributed to the development of performance measurement dashboards to track and assess the effectiveness of marketing strategies.
• Developed the technical flow/architecture and code for complex marketing campaigns, utilizing Python and SQL to streamline processes.
• Drove process enhancements by applying automation techniques, minimizing data latency, and achieving seamless data integration within the team.
• Created a data pipeline to migrate data from GCP to the Teradata database using Python and BigQuery, ensuring efficient data handling and transfer.
• Engaged with business analysts to understand business processes, the flow of data within and between processes, and recommended process improvements for better data quality and effectiveness.

• Set up docker containers on Nvidia DGX server for training deep learning models and enable rapid analytics using GPUs
• Created a dashboard to monitor server and GPU statistics using NVDashboard and deployed using Docker
• Developed a Flask Application to manage ML models and provide an admin console for NVIDIA edge devices
• Developed Python scripts for converting raw data into KITTY Format for object detection algorithms
• Enabled rapid data analytics on huge datasets using Nvidia GPU by setting up docker containers on server
• Created visualizations using Tableau Desktop and Python libraries (Plotly, Seaborn, Matplotlib)
• Schedule python scripts on Linux Servers and devices (NVIDIA Jetson Devices, NVIDIA DGX Server)
• Used transfer learning to re-train and deploy pre-trained ML models from Nvidia AI for License Plate Recognition

Data Analysis

• Developed Python scripts for cleaning and formatting data received from inbound systems
• Developed monthly sales reports in Tableau Desktop

Application Development

• Collaborated with a team of 8 members from cross-functional teams to develop PLSQL procedures and functions for data processing tailored to project requirements, ensuring alignment with business objectives.
• Scheduled shell scripts using Crontab and managed Oracle Solaris Servers based on Linux, facilitating smooth and efficient operations within the team.
• Conducted performance tuning of database procedures and SQL queries using profiling tools and index rebuild techniques, leading to significant efficiency gains.
• Enhanced application performance through strategic index rebuilds, freeing up 18GB of space on the production server and contributing to overall system stability.
• Utilized SQL Loader to efficiently load data from flat files to Oracle tables, ensuring seamless data integration for project needs.
• Worked collaboratively to develop the front-end for an Oracle Data Migration web-based tool using HTML, CSS, and JavaScript, improving user experience and accessibility for stakeholders.

International Institute of Information Technology - Bangalore (IIIT-B)

Developed practical skills in data science through industry projects, mastering tools like Python, R, MySQL, and Tableau. Gained expertise in data manipulation, analytics, and machine learning under guidance from IIIT-B experts.

View Credentials

College of Engineering Roorkee

Completed B.Tech in Electronics and Telecommunication, gaining hands-on experience in digital circuits, robotics, and technical projects while actively participating in coding challenges and cultural events.

TCG Digital

Awarded the Spot Award for delivering high-quality work on time with exceptional precision and dedication, ensuring excellent outcomes and client satisfaction.

Capgemini India

Awarded with the Project Star Award for excellence in customer delivery and outstanding performance in Oracle Apps NA Business Unit At Capgemini.

View Credentials

Nvidia Deep Learning Institute

RAPIDS is a collection of data science libraries that allows end-to-end GPU acceleration for data science workflows, and together with Dask, can leverage multiple GPUs on larger-than-memory datasets.

View Credentials