Hello, I'm Mehemud Azad

Building intelligent systems with Machine Learning, Agentic AI, and NLP

View Projects GitHub Contact

developer@portfolio:~

$ cat profile.txt

Name: Mehemud Azad

Location: Dhaka, Bangladesh

Expertise: Full Stack | Machine Learning | Agentic AI

Status: Building & Learning

$ _

01. About Me

I am a Computer Science student at Bangladesh University of Engineering and Technology, currently in my third year of studies. My journey began with a strong foundation in Full-Stack Web Development, where I gained hands-on experience building and deploying scalable web applications.

Subsequently, my focus expanded into DevOps and DevSecOps, implementing robust automation pipelines and secure cloud-native solutions. Concurrently, competitive participation in CTFs (Capture The Flag) honed my practical cybersecurity skills and specialized problem-solving capabilities.

Currently, my primary research and technical focus rests on Machine Learning, with a specialized interest in Agentic AI and Natural Language Processing (NLP) I am actively involved in innovative research projects, specifically focusing on the development of exploratory test automation frameworks for mobile environments leveraging Agentic AI, as well as advancing Bangla LLM-generated text detection methodologies through comprehensive dataset curation.

Outside of work and study, I enjoy working out and watching anime, and One Piece is my favorite. I like how it shows that big goals require patience, persistence, and creativity.

Machine Learning

DevOps

Agentic AI

NLP

LLM

CTF

02. Tech Stack

Machine Learning & AI

PyTorch Scikit-learn Transformers LangChain MLflow Optuna Kaggle

DevOps & DevSecOps

Docker Kubernetes Terraform GitHub Actions Apache Kafka DigitalOcean SonarQube Trivy

Core Languages

Python JavaScript TypeScript Java C / C++ Solidity

Backend & Databases

Node.js Spring Boot FastAPI PostgreSQL MongoDB Redis Firebase

Frontend & Tools

React / Next.js Tailwind CSS Linux Git Postman Figma

03. Featured Projects

MAZE-SOLVER-ATMEGA32 (CSE 316)

Built an autonomous maze-solving robot on ATmega32 using a left-hand rule algorithm, triple ultrasonic sensors, MPU6050 gyroscope feedback, and PID-assisted movement for reliable navigation under hardware constraints.

C++ ATmega32 PlatformIO HC-SR04 MPU6050

The Freelancer Platform

Developed a full microservices-based freelancing platform with API gateway, JWT auth, escrow payments, workspace collaboration, and AI-enhanced matching/content features powered by vector search.

Spring Boot Next.js Kafka PostgreSQL Redis

CareForAll (FAT32 API Avengers)

Champion project from CUET API Avengers: a production-ready, event-driven donation platform with FastAPI microservices, Kafka workflows, gRPC service communication, observability stack, CI/CD, and Kubernetes deployment.

FastAPI Kafka gRPC Docker Kubernetes

Green Agent - GAIA Evaluator

Implemented an A2A protocol-based evaluator for GAIA benchmark tasks using Google ADK, including deterministic scoring, optional LLM-based judging, and multi-agent orchestration for robust agent evaluation.

Python Google ADK A2A Protocol Gemini FastAPI

CSE-310 Compiler

Built a full C8086 compiler pipeline from scratch: symbol table, lexical analysis (Flex), syntax/semantic analysis (ANTLR4), and intermediate/code generation with assembly output and optimization.

C++ Flex ANTLR4 Compiler Design

04. Achievements

BUET IEEE DevSprint 2023

Rising Team

Earned my first major achievement by becoming part of a rising team, which kick-started my competitive and project-focused journey.

Teamwork Growth Foundation

CUET API Avengers 2025

Champion - API Avengers (Team FAT32)

Became champion in the CUET API Avengers DevOps competition with Team FAT32, delivering the CareForAll microservices platform with CI/CD and cloud-native deployment.

DevOps Microservices Cloud Repository

CUET Techathon 2025

Ranked 5th Place

Secured 5th position at CUET Techathon 2025, an intensive Internet of Things (IoT) based competition. Our team developed and integrated functional hardware and IoT solutions under stringent time constraints.

IoT Hardware Integration Top 5 Hackathon

BUET DL Sprint 4.0 2026

Ranked 7th out of 192 Teams

Secured 7th position among 192 teams in BUET DL-sprint-4, strengthening my practical experience in deep learning competitions.

Deep Learning Competition Top 10

05. Research Focus

Current

Agentic AI, NLP & LLMs

I am mainly focused on Agentic AI, NLP, and LLM-centered research and implementation work as my primary learning direction.

Agentic AI NLP LLMs Machine Learning

Capstone

Exploratory Test Automation with Agentic AI

In my capstone project, I am building an agentic software system for exploratory software test automation.

Capstone Agentic Systems Software Testing

Bangla NLP

Bangla LLM-Written Text Detection

I am also working on Bangla LLM-written text detection and trying to create a dataset of Bangla LLM-generated texts.

Bangla NLP LLM Detection Dataset Creation

Ongoing

Kaggle Contest Practice

In my free time I am trying participate in Kaggle contests to improve practical experimentation, feature engineering, and evaluation skills for ML workflows.

Kaggle ML Experiments Model Evaluation

06. GitHub Activity

@MehemudAzad

Building in public, one commit at a time

View Profile Repositories

Public Repos

View on GitHub

Total Stars

View on GitHub

Followers

View on GitHub

Contributions

Active Daily

07. Get In Touch

Let's Build Something Amazing

I'm always interested in hearing about new opportunities, collaborations, or just chatting about Machine Learning, NLP, Agentic AI, and software engineering. Feel free to reach out!

mehemudazad.km@gmail.com github.com/MehemudAzad LinkedIn Profile Facebook Profile

Dhaka, Bangladesh