Shiti Chowdhury Portfolio

ABOUT ME

I am a final-year CSE undergraduate at CUET with a strong passion for Artificial Intelligence, Machine Learning, NLP, Large Language Models (LLMs) and Data-Driven Innovation.

I enjoy exploring intelligent systems, performing impactful research and building real-world applications that blend creativity with engineering. I actively participate in AI/ML competitions, collaborative problem-solving and community learning initiatives. My goal is to create meaningful AI solutions that positively impact society.

Education

My academic journey reflects a strong commitment to excellence in engineering, analytical thinking and continuous learning. I have consistently worked to strengthen my understanding of software, AI/ML, research and problem-solving.

Chittagong University of Engineering & Technology (CUET)

Bachelor of Science in Computer Science and Engineering (CSE)

Final Year (Ongoing) — CGPA: 3.68 — Chattogram, Bangladesh

Currently specializing in Artificial Intelligence, Machine Learning, Deep Learning and advanced software development. Engaged in academic research and collaborative work.

Chattogram College

Higher Secondary Certificate (HSC) — Science

2020 — GPA 5.00 — Chattogram, Bangladesh

Focused on science subjects with strong analytical, mathematical and problem-solving foundations essential for engineering.

Chittagong Govt. Girls’ High School

Secondary School Certificate (SSC) — Science

2018 — GPA 5.00 — Chattogram, Bangladesh

Achieved strong academic results and developed a lifelong passion for learning, leadership and scientific exploration.

Experience

Computer Vision Analysis Intern

Brain Machine AI · Jun 2025 – Dec 2025
Chattogram, Bangladesh (Hybrid)

Worked on CCTV-based computer vision systems for monitoring and attendance analysis.
Trained and evaluated multimodal deep learning models using image, video, and audio data with Python and OpenCV.

Campus Ambassador

OmniStart 1.0 · Sep 2025 – Dec 2025
Kaizen SUST (Remote)

Selected as a Campus Ambassador representing CUET in a national startup competition.
Coordinated campus-level promotion and communication between CUET students and the organizing team.

Technical Skills

Programming & Web Technologies

Python

C++

HTML5

CSS3

JavaScript

VS Code

Django

MySQL

Bootstrap

Frameworks & Libraries

FastAPI

Streamlit

PyTorch

TensorFlow

Keras

OpenCV

Scikit-learn

NumPy

Pandas

Matplotlib

Seaborn

Plotly

HuggingFace Transformers

Data Science (AI & Machine Learning)

Machine Learning

Deep Learning

NLP

Computer Vision

Speech & Image Processing

LLMs & LangChain

Prompt Engineering

RAG

Tools & Platforms

Docker

Git

GitHub

Linux

LaTeX

Canva

Figma

MS Office

Jira

Soft Skills

Critical Thinking

Problem-Solving

Leadership

Team Collaboration

Project Management

Time Management

Initiative & Ownership

Research Publications

•

Team_Luminaries_0227 @ LT-EDI-2025: A Transformer-Based Fusion Approach to Misogyny Detection in Chinese Memes

5th Workshop on Language Technology for Equality, Diversity & Inclusion (LT-EDI 2025), September 2025
Expected Link: ACL Anthology

•

Think Tank: A Home Appliance and Vehicle Repair Assistance Smartphone Application with Technician Hiring & Repair Schedule Booking Features

2025 4th OPJU International Technology Conference (OTCON) on Smart Computing for Innovation & Advancement in Industry 5.0, July 2025
Available at: IEEE Xplore

•

Team ML_Forge @ DravidianLangTech 2025: Multimodal Hate Speech Detection in Dravidian Languages

NAACL 2025 Shared Task – Association for Computational Linguistics (ACL), May 2025
Available at: ACL Anthology

Projects

LangChain FastAPI Streamlit

Agentic AI Chatbot

Created an agentic AI chatbot using FastAPI and Streamlit with OpenAI and Groq models, supporting multi-agent reasoning workflows and optional web search integration for real-time knowledge.

🔗 View Project

Stable Diffusion Computer Vision

Synthetic Face Dataset Generation

Implemented a synthetic facial image dataset generation pipeline using Stable Diffusion with diverse emotions and demographics for computer vision and emotion recognition research.

🔗 View Project

BLIP Streamlit Computer Vision

Image Caption Generator

Built an AI-powered image captioning web application using Streamlit and a pretrained BLIP model to generate natural language descriptions from uploaded images.

🔗 View Project

Multimodal AI NLP Classification