Mar 2024 – May 2024 · 2 mos

• Developed an invoice data analyzer using PaddleOCR for PDF analysis and a suite of LLM modules for data extraction.
• Implemented coordinate analysis and LLM models for extraction of invoice details, optimizing accuracy and reliability.

Jun 2023 – Aug 2023 · 2 mos · Mumbai, Maharashtra, India · On-site

May 2022 – Jul 2022 · 2 mos

• ◦ Keyword Extraction: Extracted keywords for SEO content briefs using Google Trends and Google Ads API and python libraries- rake, yake, keybert and gensim.
• ◦ Ranking Algorithm: Designed and implemented a ranking algorithm for the extracted keywords which ranks them based on several factors like relevance to the search topic(Cosine Similarity), interest over time data, inverse document frequency, etc.
• ◦ Code Optimisation: Optimised and parallelized(using python libraries like threading, multiprocessing and joblib) the extraction and ranking algorithms to reduce the execution time of the queries by 50%.
• ◦ Content Brief Recommendations: Implemented an algorithm for word, para, heading and keyword count recommendations.

Dec 2021 – Feb 2022 · 2 mos

• ◦ Designed and Implemented a Retool application in JavaScript that generates Skyscraper, Futoshiki and
• sudoku puzzles in JSON format by taking in several inputs – size of matrix, number of empty cells, duration, score, etc.
• ◦ Scraped sudoku puzzle sites using the python library Scrapy.