Oct 2023 – Aug 2024 · 10 mos · Bengaluru, Karnataka, India · Hybrid

• Designed and implemented auto-scalable backend systems & ETL pipelines integrating social media &
• e-commerce platforms using Airflow, SQL, Python.
• Optimized Odoo ERP system by building a custom SQL query engine, reducing query latency by 90%.

Apr 2022 – Oct 2023 · 1 yr 6 mos · India

• Created an AI-powered Bank Statement Analyzer using Object Detection & NLP, extracting 700+ financial
• features for risk assessment and loan underwriting using multi-modal transformers, Named Entity Recognition (NER), and OCR.
• Engineered a scalable document processing pipeline, automating BNPL (Buy Now Pay Later) & Insurance verification services, processing 100,000+ documents/day.

Oct 2021 – Apr 2022 · 6 mos · Banglore

• Engineering large scale handwriting recognition model of bank forms for various banks across the world using custom models using Time Distributed CRNN model with 89% accuracy of predictions to facilitate the digitization of banking forms.
• Modelled and Regulated Invoice Extraction through OCR character recognition to extract details like goods and description, GST, and order number.

Jul 2021 – Oct 2021 · 3 mos · Ahmedabad, Gujarat, India

• Coordinated with Scientists and examined large scale high-resolution satellite hyperspectral and multispectral images by co-registration the data to find unique characteristics of the land and classify based on ground truth like water, forest, concrete structures etc using an ensemble machine learning model with a prediction accuracy of 98.8%.
• The final results were used to develop a plugin for ENVI for image analysis, ISRO’s official remote sensor Analysis tool. Developed various ensemble-based Machine learning models and various Deep learning ensemble models for predicting the plan based on it intensity characteristics of the image on the electromagnetic spectrum. We used Pixel intensity-based as well as OBIA(Object-based intensity Analysis) for developing machine learning models. Achieved a testing Accuracy of 98.3% on seven different classes of ground truth using pixel intensity-based algorithms upon testing.

Apr 2021 – Jul 2021 · 3 mos · Leamington Spa, England, United Kingdom

• Engineered various Proof of Concept(POC) projects for clients, giving them the better insights about the solution proposed by the
• company to solve their existing business problem.
• The Client’s solution satisfaction rate was increased by 14% after presenting POC Projects.

Oct 2020 – Jun 2021 · 8 mos · Gandhinagar, Gujarat, India

Jan 2020 – Dec 2020 · 11 mos · India

• Conducted learning sessions amongst 60+ peers and faculties of the university to teach concepts statistical learning and implementations in different domains like Engineering, Finance, Economics, and Business helping them in projects and publications.