Quamrul Hoda

Quamrul Hoda

|

AI Engineer focused on building production-grade Agentic AI systems, RAG pipelines, multi-agent workflows, and intelligent automation solutions that solve real-world problems.

100+ DSA Problems Solved
9+ AI Projects Built
1 Internships
7.2 CGPA
View Projects Let's Connect View Resume

Get to Know Me

About Me

I'm Quamrul Hoda, a B.Tech student specializing in Artificial Intelligence and Machine Learning at IK Gujral Punjab Technical University with a CGPA of 7.1/10.

During my internship at CodeAlpha, I worked on restaurant price prediction systems, anomaly detection pipelines, and applied ML techniques across Python, Scikit-learn, TensorFlow, and Pandas — delivering robust, production-ready solutions.

I'm deeply passionate about the convergence of NLP, LLMs, and Agentic AI. My projects span RAG-based multi-agent systems, YouTube comment sentiment analysis, and medical image classification — each pushing me to design smarter, more reliable AI systems.

I thrive at the intersection of research and engineering — turning complex ML concepts into well-structured, tested, and evaluated systems ready for the real world.

Current Focus

Agentic AI RAG Systems LLMs & NLP MLOps Deep Learning FastAPI LangChain LangGraph Vector Databases
qhoda489@gmail.com +91-7851819968

My Education

Education Is Not The Learning Of Facts, But The Training Of The Mind To Think.

IK Gujral Punjab Technical University

Bachelor Of Engineering In Computer Science And Engineering

Ik Gujral Punjab Technical University | Jalandhar Punjab

CGPA: 7.2 / 10 2022 | Pursuing
KV Bailey Road Patna

Intermediate In Science

D.S College Katihar | BSEB

Percentage: 72%

What I Work With

Technical Expertise

Languages & Async

Python
Async Programming
SQL

AI & NLP

Transformer Architecture
LLMs & Prompt Engineering
Embeddings & RAG Systems
NLP & Fine-Tuning
LLM Evaluation

Agentic & Multi-Agent AI

LangChain / LangGraph
Agentic Workflows
Context Engineering
OpenAI API

ML/DL Frameworks

Scikit-learn
TensorFlow / Keras
PyTorch
XGBoost / LightGBM
HuggingFace Transformers

MLOps & Tools

FastAPI
Docker
AWS Cloud
GitHub / CI/CD Pipelines
MLflow / DVC
Unit Testing

Databases & Vector Stores

PostgreSQL
MongoDB
FAISS (Vector Store)
Redis
Pinecone / Chroma

My Journey

Work Experience

Machine Learning Intern

CodeAlpha

1st — 30th July 2025 Remote
  • Designed and deployed a Restaurant Price Prediction system using supervised ML algorithms, building an end-to-end retrieval and evaluation pipeline for structured output and model reliability.
  • Applied CatBoost and anomaly detection to handle high-cardinality features and identify outliers, improving system correctness and prediction robustness.
  • Implemented ML pipelines using Python, Scikit-learn, and Pandas covering data ingestion, feature engineering, model training, and evaluation frameworks.
Domain Supervised ML
Type Remote Internship
Duration 1 Month
Python Scikit-learn CatBoost Pandas Anomaly Detection Evaluation Frameworks

What I've Built

Featured Projects

AI Chatbot RAG Multi-Agent System
Agentic AI / NLP

AI Chatbot — RAG Multi-Agent System

Production-grade agentic chatbot with LangGraph workflows, FAISS RAG retrieval, and session management via MongoDB.

  • RAG retrieval pipeline with FAISS & OpenAI API
  • Async FastAPI backend with streaming responses
🎯 Reduced irrelevant outputs by 40% via semantic search
LangGraph FastAPI FAISS OpenAI MongoDB
YouTube Comment Analysis NLP Pipeline
NLP & Evaluation

YouTube Comment Analysis — NLP Pipeline

Real-time comment sentiment and toxicity classification using BERT Transformer architecture with sub-second latency.

  • Fine-tuned BERT Transformers for classification
  • FastAPI inference API with GitHub Actions CI/CD
🎯 98% accuracy on domain-specific comment data
BERT HuggingFace FastAPI PyTorch CI/CD
Kidney Disease Classification Deep Learning
Deep Learning / Medical AI

Kidney Disease Classification

Medical image classification using CNN transfer learning (VGG16) with an MLOps pipeline for clinical datasets.

  • CNN with VGG16 transfer learning
  • MLOps pipeline with unit-tested ingestion
🎯 88% diagnostic accuracy on clinical CT scans
CNN VGG16 TensorFlow Keras MLOps
Swiggy Delivery Time Prediction System
Machine Learning / MLOps

Swiggy Delivery Time Prediction

End-to-end regression pipeline for estimating food delivery durations based on distance, traffic, and order metrics.

  • Feature engineering with LightGBM & FastAPI
  • Automated data ingestion and prediction pipeline
🎯 96% accuracy on delivery duration test dataset
Python LightGBM FastAPI Scikit-Learn
Wine Quality ML Project
Supervised ML / Analytics

Wine Quality Prediction

Supervised classification system evaluating physicochemical properties to predict wine quality grades.

  • Classification pipeline with LightGBM & FastAPI
  • Outlier filtering & feature scaling pipeline
🎯 93% accuracy on physicochemical quality metrics
Python LightGBM FastAPI Scikit-Learn

Credentials

Certifications

Industry-recognized credentials in AI, NLP, Agentic AI, and RAG.

Agentic AI using Agno Certificate
View Certificate
Agentic AI using Agno
Advanced RAG Certificate
View Certificate
Advanced RAG
Data Science with Generative AI Certificate
View Certificate
Data Science with Generative AI

Let's Build Together

Ready to collaborate on AI projects or discuss opportunities in agentic AI and production GenAI systems.

Message sent! I'll get back to you soon.
Quamrul's AI Assistant