Available for work

MD Abdullah
Ibne Aziz

I build AI that actually ships. Computer vision, LLM agents, and production ML pipelines, engineered to hold up outside the notebook.

Large Language ModelsRAG SystemsAI AgentsMLOpsComputer VisionPythonPyTorchNumPyPandasFastAPIPostgreSQLDockerNext.jsReactFlaskLarge Language ModelsRAG SystemsAI AgentsMLOpsComputer VisionPythonPyTorchNumPyPandasFastAPIPostgreSQLDockerNext.jsReactFlask
About Dhaka, Bangladesh

I'm a machine learning engineer drawn to the messy middle: taking an idea from a research notebook all the way to a system in production. My range runs from computer vision pipelines to LLM-powered automation.

Currently a Jr. Artificial Intelligence Engineer at AIQNIC and a Computer Science & Engineering graduate from IUBAT. I specialize in designing intelligent systems using machine learning, computer vision, and large language models, with experience building AI solutions from research and prototyping to real-world applications.

2+ Years experience
20+ AI solutions built
3 AI domains · CV · NLP · LLMs
Skills & Expertise Full-stack AI engineering
Experience Where I've worked
2026 to Now
Jr. Artificial Intelligence Engineer AIQNIC

Building production AI systems: computer-vision pipelines, LLM-driven workflow automation, and interpretable deep-learning models, taking research prototypes from the notebook into shipped products.

2022 to 2026
BSc, Computer Science & Engineering IUBAT, International University of Business Agriculture and Technology

Developed a strong foundation in computer science, software engineering, algorithms, databases, artificial intelligence, and machine learning through academic coursework and practical projects.

Selected Work 2024 to 2026
01
Smart Eye: AI Car Wash Monitoring

Desktop app with a custom YOLO11 detector (99.49% mAP@50), ByteTrack and Re-ID tracking for stable vehicle IDs through occlusion, automated session logging, capacity alerts, and analytics, built for a car wash from the client's own CCTV footage.

YOLOv11ByteTrackSQLiteOpenCV
Computer Vision ↗
02
AI-Powered CV Screening & Ranking

An n8n automation pipeline that pulls CVs from Google Drive, OCRs scanned PDFs with Gemini 2.0 Flash, scores candidates with GPT-4o against weighted job criteria, and ranks them into Google Sheets, cutting manual screening time by ~90%.

n8nGPT-4oGemini 2.0Workflow Automation
LLM Automation ↗
03
Product Inventory Management System

A web-based inventory platform for Fold Fashion BD that tracks products, sales, and stock across multiple branches, with per-branch stock levels, overselling validation, automatic stock adjustments on every transaction, and REST APIs backed by transaction-safe PHP/MySQL.

PHPMySQLREST API
Full-Stack Web ↗
04
Real-Time AI Analysis for SMEs

A branch-based sales & inventory platform for small and medium enterprises, with role-based Admin and Staff workspaces, multi-branch management, and an AI forecasting layer that predicts demand and flags shortfalls through real-time dashboards and PDF reports.

AI ForecastingRole-basedMulti-branch
AI Analytics ↗
Publications Peer-reviewed research
2026
Interpretable Diabetic Retinopathy Detection and Grading Using EfficientNet-ViT Hybrid Architecture with Visual Lesion-Level Grad-CAM Analysis ICECTE 2026 · IEEE · doi:10.1109/ICECTE69292.2026.11429403
2026
EffNet-ViT: An Interpretable Hybrid EfficientNet-Transformer Architecture for Ischemic Stroke Classification on CT Scans ICECTE 2026 · IEEE · doi:10.1109/ICECTE69292.2026.11429454
2025
A Deep Learning Approach for Detecting Pests and Diseases in Maize Crops Springer Nature · CCIS · doi:10.1007/978-3-032-11335-1_23
2025
Deep Learning-Based Classification of Ischemic Stroke Using Brain CT Scans IDAA 2025 · International Conference on Intelligent Data Analysis and Applications · doi:10.2991/978-94-6239-664-7_43
2025
A Comparative Study of Transfer Learning Models for Cotton Plant Anomaly Identification with Explainable AI RAAICON 2025 · IEEE · doi:10.1109/RAAICON69033.2025.11502311
2025
A Machine Learning Approach for Predicting Academic Performance Using Classification Models ICCCT 2025 · Springer · doi:10.1007/978-981-95-3498-2_30
Portrait
Get in touch

Let's build something intelligent.

Open to AI engineering roles, research collaborations, and freelance ML projects. I usually reply within a day.

Email aziz.miraz123@gmail.com GitHub AbdullahIbneAziz LinkedIn MD Abdullah Ibne Aziz