Joel Franklin Stalin Vijayakumar
Machine Learning Engineer @Emerjence Llc
Signup · Get unlimited contacts
WORK HISTORY
Machine Learning Engineer @Emerjence Llc
Ashburn, VA, US
Developed a proof-of-concept (POC) Retrieval-Augmented Generation (RAG)-based chatbot using Streamlit that allows users to upload multiple PDF documents and interact with their content through natural language queries. Indexed documents using OpenAI Embeddings and ingested them into the Pinecone vector store for efficient retrieval. Leveraged LangChain to rephrase user queries, preserve chat history and fetch relevant context, which was processed by GPT-4o LLM to produce accurate, context-aware responses. Reduced document search and response time by 40%. Engineered a POC Excel Worksheet Analyzer leveraging LangChain and OpenAI APIs (GPT-4) in a multi-agent architecture capable of intelligently generating and executing Python code for complex Excel data analysis via intuitive, no-code natural language queries, reducing manual analysis effort by 45%. Architected and deployed a full-stack serverless knowledge management system using React/TypeScript frontend and Hono/Cloudflare Workers backend with PostgreSQL/Prisma database delivering sub-10ms cold starts. Developed an end-to-end application using AWS Lambda, Amazon Bedrock, and AWS API Gateway to automate blog content generation. Designed robust API endpoints to trigger Lambda functions, utilizing Boto3 for invoking foundational models (Llama 2 and Nova) on Amazon Bedrock, and seamlessly storing generated content as text files in Amazon S3, boosting content creation efficiency by 55% and reducing manual intervention by 60%. Engineered an order assistant chatbot using Microsoft Copilot Studio, integrating it with Quickbase database to fetch menu options and utilizing a UiPath Robotic Process Automation workflow running on an Azure Virtual Machine (VM) to record order details in a CSV file and produce order numbers for customers. Minimized order processing time by 30% and improved order accuracy by 25%.
EDUCATION
D.A.V Senior Secondary School
12th Standard
Boston University Graduate School of Arts & Sciences
Master's degree, Computer Science (Specialization in Machine Learning)
Birla Institute of Technology and Science, Pilani
B.E (hons) , Mechanical Engineering
D.A.V Senior Secondary School
10th Standard
ABOUT JOEL FRANKLIN STALIN VIJAYAKUMAR
I am at the forefront of Machine Learning (ML) and Generative AI (GenAI) at Emerjence, where I specialize in Retrieval-Augmented Generation (RAG) systems and Robotic Process Automation (RPA). My professional journey spans organizations like IISc Bangalore and Kissht, where I have worked extensively on cutting-edge GenAI and NLP technologies. With a Master’s in Computer Science (specializing in ML) from Boston University and a Bachelor of Engineering from BITS Pilani, my academic and professional experiences have seamlessly aligned with the advancements in NLP and GenAI. My journey into GenAI has been anything but abrupt. NLP is widely regarded as one of the foundational pillars of GenAI, and I have been deeply engaged in NLP since 2020, witnessing and contributing to its evolution into what is now recognized as GenAI. I have strong fundamentals in NLP and GenAI. I have worked on diverse professional and personal projects which include developing an Automatic Speech Recognition (ASR) system by fine-tuning the Wav2Vec 2.0 model, building a Neural Machine Translation (NMT) model leveraging T5\'s encoder-decoder architecture, designing a Trigger Word Detection system using GRU architecture and fine-tuning BERT and RNN architectures for Sentiment Analysis.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.