About Research Experience Projects Skills Contact
Open to Global Opportunities

Harshal Dharpure

AI Researcher ML Engineer NLP Specialist

M.Tech in Artificial Intelligence @ IIT Patna

Specialized in Large Language Models, Retrieval-Augmented Generation, and Multimodal AI. Building next-generation AI systems with research experience at premier Indian Institutes of Technology.

2 IIT Research
6+ Awards
🏆 SIH Winner
ai_researcher.py
class AIResearcher:
    def __init__(self):
        self.name = "Harshal Dharpure"
        self.role = "AI Researcher & ML Engineer"
        self.education = "M.Tech AI @ IIT Patna"
        self.expertise = [
            "Large Language Models",
            "RAG Systems",
            "Multimodal AI",
            "NLP & Transformers",
            "MLOps & Deployment"
        ]
    
    def solve(self, problem):
        return "Innovative AI Solution"

01. About Me

I'm an AI Researcher and Machine Learning Engineer passionate about pushing the boundaries of artificial intelligence.

Currently pursuing my M.Tech in Artificial Intelligence at the Indian Institute of Technology, Patna, I specialize in building intelligent systems that leverage the power of Large Language Models, Retrieval-Augmented Generation (RAG), and Multimodal AI.

My research experience spans two premier IITs — at IIT Patna, I worked on multimodal persuasive content detection using LLMs, and at IIT Gandhinagar, I conducted research on bias detection in transformer models. This unique blend of research excellence and hands-on engineering experience positions me to tackle complex AI challenges in both academic and industry settings.

I'm actively seeking global opportunities in AI research, machine learning engineering, and NLP roles where I can contribute to cutting-edge AI development.

Education

Current

M.Tech in Artificial Intelligence

IIT Patna • 2025-Present

80.03%

B.E. in Computer Science

Sipna College of Engineering • 2019-2023

Research Interests

Large Language Models Retrieval-Augmented Generation Multimodal Learning NLP & Transformers AI Safety & Bias Prompt Engineering

02. Research Experience

IIT Gandhinagar

Gender Bias Detection in Language Models

SRIP Research Intern • May 2021 - Jul 2021

Investigated gender bias in multilingual transformer models using association tests and fairness metrics on code-mixed text.

  • Conducted WEAT (Word Embedding Association Test) on mBERT embeddings
  • Implemented HONEST score computation for bias measurement
  • Analyzed bias patterns across multiple languages using Hugging Face
mBERTBias DetectionNLPHugging Face

03. Professional Experience

Teaching Assistant

IIT Patna

Aug 2025 - Present
  • CS1101 – Foundation of Programming under guidance of Dr. Arijit Mondal
  • Assisting in tutorials, lab sessions, and evaluations for foundational programming course
  • Mentoring undergraduate students in programming concepts and problem-solving
TeachingProgrammingMentoring

Product Engineer (DevOps)

Beacon Healthcare Systems

Jul 2024 - Aug 2025 • 1 yr 2 mos
  • Led design and implementation of scalable infrastructure for enterprise healthcare applications
  • Specialized in HAProxy reverse proxy configuration for secure API routing across multi-client environments
  • Automated deployments using Ansible with standardized roles and playbooks
  • Managed JasperReports Server on Tomcat with JVM tuning for QA/PROD environments
  • Designed reusable base VM templates and internal tooling for rapid client onboarding
HAProxyAnsibleMySQLLinuxTomcatCI/CD

Jr. Product Engineer

Beacon Healthcare Systems

May 2024 - Jun 2024 • 2 mos
  • Contributed to MySQL and Tomcat deployments with server hardening
  • Created comprehensive deployment documentation and operational runbooks
  • Supported DevOps team in client environment setup and proxy configuration
MySQLTomcatLinuxDocumentation

AWS Cloud Foundations Intern

Amazon Web Services (AWS)

Oct 2021 - Dec 2021 • 3 mos
  • AWS Academy Graduate - AWS Academy Cloud Foundations Program
  • Hands-on experience with EC2 instance provisioning, configuration, and management
  • Worked with S3, IAM, and VPC, building foundation in cloud infrastructure and security
AWSEC2S3IAMVPC

04. Featured Projects

AI Application

Intelligent Document Q&A with RAG

Production-ready RAG system for document question-answering using LangChain, vector databases, and fine-tuned embedding models.

  • Supports PDF, DOCX, and web content
  • ChromaDB vector store with semantic search
  • Streaming responses with source citations
LangChainRAGChromaDBOpenAI
NLP Project

Multi-Turn Conversational AI Chatbot

Context-aware chatbot with memory management, intent classification, and entity extraction for domain-specific conversations.

  • Fine-tuned DialoGPT for domain adaptation
  • Conversation memory with summarization
  • REST API with WebSocket support
TransformersDialoGPTFastAPIRedis
NLP Research

Sentiment Analysis with Aspect Extraction

Fine-grained sentiment analysis system that extracts aspects and determines sentiment polarity for product reviews.

  • BERT-based aspect term extraction
  • Multi-class sentiment classification
  • Deployed on AWS with auto-scaling
BERTNLPAWSDocker
Computer Vision

Image Captioning with Visual Attention

Deep learning model for generating descriptive captions for images using attention mechanisms and encoder-decoder architecture.

  • ResNet-101 encoder with attention
  • LSTM decoder with beam search
  • BLEU-4 score of 32.5 on COCO
PyTorchCNNLSTMAttention
MLOps

End-to-End ML Pipeline with MLflow

Automated machine learning pipeline with experiment tracking, model versioning, and deployment automation.

  • Automated hyperparameter tuning
  • Model registry with A/B testing
  • Kubernetes deployment with monitoring
MLflowKubernetesDockerGitHub Actions

05. Technical Arsenal

AI/ML & Deep Learning

Large Language Models RAG Systems Transformers PyTorch TensorFlow Hugging Face LangChain Fine-tuning

NLP & Computer Vision

NLP Multimodal AI BERT/GPT CLIP/BLIP Prompt Engineering CNN Attention Mechanisms

Programming & Tools

Python C/C++ SQL Git Linux FastAPI Streamlit

MLOps & Cloud

Docker Kubernetes MLflow AWS CI/CD GitHub Actions Ansible

06. Honors & Awards

Top 10%

AWS Deepracer League 2022

August Qualifier • Global Open Division

Top 20 Semi-finalist

Adobe Analytics Challenge 2021

4000+ Teams Globally • Real-world Data

Most Popular Team

Sparkathon 2022

National Level 24-Hour Hackathon

Runner Up

Code the Web 2022

Web Development Competition

Diary Writing Award

e-SRIP 2021

IIT Gandhinagar

07. Organizations & Memberships

Computer Society of India

Member • Sep 2020 - May 2023

National body representing computer professionals with 72 chapters across India

LINGO Research Group

Research Intern • IIT Gandhinagar

Computational Linguistics and Complex Social Networks Group

08. Let's Connect

I'm actively seeking global opportunities in AI research, machine learning engineering, and NLP roles. Whether you're a recruiter, researcher, or fellow AI enthusiast, I'd love to hear from you!

Get In Touch