Hi, I am

Nihaarika Agarwal

MS Computer Science

I'm a Computer Science MS student at Arizona State University, specializing in AI. I’m passionate about using machine learning, NLP, and generative AI to solve real-world problems.

My experience includes research and development of RAG systems with VectorDB, prompt engineering, and RPA solutions that drive efficiency and impact. I enjoy building practical, trustworthy AI systems and constantly seek opportunities to learn, collaborate, and innovate.

Outside tech, I love reading and writing short poems. Let’s connect if you have an idea or challenge that could use a creative, analytical mind!

Resume

My Tech Toolkit

Tools/Libraries

ML & Data Science Frameworks

TensorFlow

PyTorch

Pandas

Matplotlib

Scikit-learn

Excel

Apache Spark

MATLAB

LLM, RAG and VectorDB

OpenAI

Hugging Face

LangChain

LangGraph

LLAMA

Mistral

Qdrant

FAISS

ChromaDB

Automation Tools

UiPath

Automation Anywhere

Jira

DevOps and Deployment

AWS

GitHub

Docker

CUDA

Flask

Streamlit

Programming Languages

Python

VBScript

Java

SQL

C++

Research domains/Skills

Machine Learning, Natural Language Processing, Large Language Model (LLM), AI Architecture, Deep learning, Retrieval Augmented Generation (RAG), Generative AI, Prompt Engineering, Human ML Interaction, Robotic Process Automation.

Professional Experience

Research Assistant

Arizona State University, Tempe, US

September 2024 – Present

LLMRAGQualitative Analysis VectorDBPrompt EngineeringHuman-ML Interaction

Researcher @ Happy (Human Aspects in Cyber Protection and Security) Lab under Prof. Jaron Mink in collaboration with SEFCOM
Improved IRB feedback quality by enabling early detection of design risks—built an LLM-powered ethics compliance tool using RAG and VectorDB.
Uncovered key prompt-response patterns in cybersecurity education through qualitative analysis of LLM outputs.
Enhanced LLM evaluation by designing prompt-analysis workflows to assess accuracy, clarity, and educational value.

Consultant

Deloitte Touche Tohmatsu, India

June 2024 – August 2024

UiPathRPAVB ScriptJIRA MS Excel MacrosAgile Methodologies

Reduced bot debugging time by 60% by designing a VBScript-based dashboard to monitor 20+ production bots.
Accelerated RPA scaling across global clients by rapidly integrating custom error-reporting modules.
Streamlined production handoffs by creating comprehensive SOPs for cross-functional teams.

Analyst

Deloitte Touche Tohmatsu, India

September 2022 – May 2024

UiPathRPAVB SriptJIRA MS Excel MacrosAgile Methodologies

Led to ~30% resource optimization for a leading global consumer healthcare supply chain by designing and standardizing automation workflows.
Resolved high-priority incidents across global units while maintaining 20+ production bots.
Improved project tracking and delivery alignment by creating JIRA pipelines and drafting client-ready documentation.
Twice awarded Deloitte’s “Live the Dot” award for innovation and measurable client impact.

Intern

Deloitte Touche Tohmatsu, India

January 2022 – June 2022

RPAAutomation AnywhereUiPathBlue Prism SDLC

Reduced turn around time by 95% for a finance workflow by developing an RPA solution collaboration with domain experts.
Trained on UiPath, BluePrism, and Automation Anywhere; developed automation solutions simulating core SDLC stages.

Undergraduate Research Assistant

Manipal Institute of technology, Manipal, India

July 2021 – December 2021

PythonNLPNLTKPandas Matplotlib

Published research on sentiment analysis using NLTK to evaluate pandemic-related mental health trends; processed large unstructured text datasets and built classifiers to detect emotional patterns during COVID-19.

Intern

Telecommunications Centre, govt of India, India

June 2021 – August 2021

PythonMLAI Fairness

Co-authored research on fairness certification for AI- proposed a “Fairness Score” framework now cited in ongoing regulatory discussions.
Conducted bias audits using Python, ML libraries; developed methods to assess algorithmic fairness in public-facing systems.Published research on sentiment analysis using NLTK to evaluate pandemic-related mental health trends; processed large unstructured text datasets and built classifiers to detect emotional patterns during COVID-19.

Projects

Corrective RAG

April 2025 – May 2025

Self-correcting RAG using LangGraph, LangChain, HuggingFace, and ChromaDB with LLM grading and query rewriting.

Cloud Based Face Recognition Pipeline AWS

Jan 2025 – May 2025

Built serverless AWS Greengrass face recognition pipeline using MTCNN, InceptionResNetV1, Lambda, SQS, for video frames.

Renewable Energy Assistant LLM Powered Streamlit App

Apr 2025 – May 2025

Built a Streamlit app using Flan-T5, Qdrant, and LangChain for time-series solar analysis and policy Q&A via RAG.

Evaluating VectorDB Augmented LLMs

Aug 2024 – Dec 2024

Evaluated generated plan accuracy in two scenarios using LLMs' multimodal capabilities with VectorDB, RAG, LLAMA, FAISS & Python.

Traffic Time Series Analysis

Dec 2024 – Jan 2025

Analyzes and forecasts traffic patterns using time series, data processing, machine learning, and statistical modeling in python.

More Projects

Published Works

Covid-19 Impact on Mental Health: Sentiment Analysis Using NLTK

6th International Conference on Contemporary Computing and Informatics(IC3I Conference) · pp. 850–855

View Publication

Fairness Score and Process Standardization: Framework for Fairness Certification in Artificial Intelligence Systems

AI Ethics Journal · Vol 3 · pp. 267–279

View Publication

Want to discover more!

Send me an email or contact
via
the instant message!

nihaarika.a22@gmail.com