Amrutha T.

Key Skills

Software

No software listed

Top Subject Matter

Generative AI

retrieval-augmented generation (RAG)

document Q&A

Top Data Types

Document

Text

Top Task Types

Question Answering

Classification

Freelancer Overview

Full Stack Data Science & GenAI Trainee - Naresh i Technologies. Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include LangChain, LangGraph, and OpenAI API. Education includes In Progress, Naresh i Technologies (2025) and Master of Science, Indiana Institute of Technology (2025). AI-training focus includes data types such as Document and Text and labeling workflows including Question Answering and Classification.

Labeling Experience

Full Stack Data Science & GenAI Trainee - Naresh i Technologies

TextClassification

Worked as an end-to-end Data Science and GenAI trainee focusing on data analysis, model building, and LLM application development. Cleaned and explored datasets in Python to derive actionable business insights while comparing multiple supervised learning approaches. Built GenAI features using LLMs and retrieval-augmented generation with LangChain, including NLP preprocessing and semantic search. • Perform EDA and data cleaning with Pandas, NumPy, Matplotlib, and Seaborn • Develop and evaluate classification and regression models across multiple domains • Implement RAG-based chatbots and document Q&A systems with LangChain • Build Streamlit frontends and apply NLP techniques for embeddings and semantic retrieval

2025 - Present

Full Stack Data Science & Gen AI Trainee (Naresh i Technologies) — Dec 2025 – Present

DocumentQuestion Answering

Developed and evaluated LLM-grounded RAG pipelines for chatbots and document Q&A, including preprocessing, embedding generation, and semantic search. Focused on reducing hallucinations by constraining generation to retrieved context and providing source citations in an interactive UI. Worked on end-to-end GenAI application development and deployment progress using Python and NLP tooling. • Built RAG-based question answering over research papers with chunking, overlap, and top-k retrieval using sentence-transformers and ChromaDB. • Generalized the RAG approach to handle arbitrary user-uploaded PDFs with variable formatting and extraction challenges. • Applied NLP preprocessing and embedding generation to support semantic retrieval and search. • Implemented and iterated on LLM pipelines using OpenAI/Gemini/Ollama with LangChain/RAG tooling, preparing components for cloud deployment.

2025 - Present

Education

Work History