About Skills Experience Projects Awards Education Contact

IDENTITY CONFIRMED_

NIKUNJSHARMA

RESEARCH INTERN @ IIT MADRAS CGPA 8.34 / 10

AI & Data Science Student

// IIIT RAICHUR — B.TECH 2023–2027

AI & Data Science undergraduate with hands-on experience in Generative AI, LLMs, NLP, and end-to-end ML systems. Currently doing research at WSAI-IIT Madras on Multilingual Indic Text Embedding using Contrastive Learning.

View Projects Contact Me

SCROLL

// 01 — IDENTITY FILE

ABOUT ME

// NIKUNJ SHARMA

With great power comes great responsibility — and in AI, that power is data. I'm an AI & Data Science undergraduate at IIIT Raichur, currently serving as a Research Intern at the WSAI Lab, IIT Madras, focusing on Multilingual Indic Text Embedding using Contrastive Learning.

Proficient in Python, SQL, REST APIs, and building end-to-end AI-driven solutions across the full ML lifecycle — from data preprocessing to model deployment. I build things that matter: NLP systems, computer vision pipelines, and full-stack platforms.

GATE-DA-2026 AIR 1477. Top 70 from 20,000+ in AlgoUniversity Fellowship Test. Strong in Data Science, ML, DBMS, DSA, OOP, and software engineering principles.

8.34

CGPA

Projects

1477

GATE AIR

// 02 — TECH ARSENAL

SKILLS

Languages

Python

JavaScript

C / C++

SQL

Dart / Flutter

AI / ML / NLP

LLMs & RAG

Prompt Engineering

Transformers / HuggingFace

Embeddings & Semantic Search

XGBoost / Scikit-learn

TensorFlow / PyTorch

NER / Regex NLP

ELO Rating Systems

Frameworks & Tools

Streamlit / Plotly

Groq SDK / asyncio

React / Node.js

OpenCV

Git / REST APIs

SQLite / MongoDB / MySQL

Pandas / NumPy / EDA

HuggingFace Spaces

// 03 — FIELD OPERATIONS

EXPERIENCE

MAY 2026 — PRESENT

Research Intern

WSAI Lab, IIT Madras

Benchmarking multilingual embedding models (IndicBERT, MuRIL, IndicSBERT, Vyakyarth) for semantic similarity and retrieval across Indic languages.
Applying contrastive learning, semantic alignment, and cross-lingual retrieval techniques; evaluating quality via cosine similarity, clustering, and retrieval-based benchmarks.
Building NLP evaluation pipelines using Python, transformer encoders, vector embeddings, and Jupyter notebooks.

// 04 — CLASSIFIED MISSIONS

PROJECTS

GEN-BI

ONGOING

Gen-BI: LLM-Powered Business Intelligence

AI analytics system converting natural language queries to SQL for enterprise databases using LLMs with RAG-based retrieval. Benchmarking multiple LLMs on SQL generation accuracy and latency.

LLMsRAGSQLPythonNLP

ARENA

⬤ LIVE

AutoEval Arena — LLM Evaluation Platform

End-to-end LLM benchmarking with async parallel inference, LLM-as-judge ensemble scoring (GPT-OSS 120B, 3-run, temp=0), chess-style ELO ratings across 6 head-to-head matchups & a self-improving agent that auto-generates harder prompts targeting model weak spots.

LLMsGroq SDKELOasyncioStreamlitPlotlySQLiteHuggingFace

Live Demo GitHub

FINSIGHT

ML PIPELINE

Finsight — Credit Risk & Churn Analysis

End-to-end ML pipeline for credit risk assessment and churn prediction using XGBoost. EDA to uncover trends, correlations, and risk indicators. Evaluated via accuracy and ROC-AUC.

XGBoostPythonEDASklearn

GitHub

OMR

COMPUTER VISION

OMR Detection & Automated Evaluation

Automated OMR sheet evaluation using OpenCV — thresholding, contour detection, region extraction — achieving ~90% accuracy through optimized preprocessing.

OpenCVPythonComputer Vision

GitHub

L&F

FULL STACK

Lost & Found Web Platform

Full-stack platform with React, Node.js, and MongoDB. JWT authentication, RESTful APIs for item reporting and recovery, responsive UI with role-based access for students and admins.

ReactNode.jsMongoDBJWT

GitHub

// 05 — COMMENDATIONS

ACHIEVEMENTS

AIR 1477

GATE DA 2026

All India Rank 1477 in GATE Data Science & AI on the very first attempt — nationwide competitive exam.

TOP 70

AlgoUniversity Fellow Test 2024

Selected among top 70 candidates from 20,000+ participants nationwide in the ATF 2024 test.

8.34

Academic Excellence

Maintained CGPA of 8.34/10 throughout the B.Tech program at IIIT Raichur.

T&P

Corporate Relations Member

Training & Placement Cell, IIIT Raichur — facilitating industry-academia collaboration.

// 06 — TRAINING LOGS

EDUCATION

2023 — 2027

B.Tech — AI & Data Science

IIIT Raichur, Karnataka

CGPA: 8.34 / 10

Specializing in AI, ML, NLP, and Data Science. Active in research, hackathons, and the Training & Placement Cell as Corporate Relations Member.

// 07 — OPEN CHANNEL

CONTACT

The signal is always on. Research opportunities, project collaborations, internship offers — reach out and let's build something.

nikunjsharma17rtm@gmail.com