As an expert in Computer Vision and Large Language Models (LLMs), I bring over six years of experience in software engineering, specializing in developing advanced AI-driven solutions. My technical prowess spans deep learning frameworks like PyTorch and TensorFlow, coupled with hands-on experience in computer vision tools such as OpenCV. In parallel, my extensive background in software development using Ruby on Rails and PostgreSQL enables me to design scalable, efficient, and data-driven applications.

I have successfully led research projects focused on image recognition, object detection, and retrieval-augmented generation (RAG) techniques, improving AI systems' accuracy and relevance. My work has been featured in peer-reviewed journals, showcasing innovations in automated damage detection and AI model optimization. By leveraging my cross-disciplinary expertise in AI and software development, I am committed to driving impactful technological advancements and delivering solutions that bridge the gap between research and practical applications.

• Master of Science in Artificial Intelligence - University of Huddersfield, UK (2022 - 2023)
  • Awarded Distinction for thesis on "Damaged Pallet Racking Classification Using Attention CNN", contributing to a 15% improvement in detection accuracy over existing methods.
• Bachelor of Science in Computer Engineering - BRAC University, Bangladesh (2013 - 2017)
  • Awarded Dean's List; Received Vice Chancellor Award for thesis project on "Handwritten Digit Recognition Using Convolutional Neural Networks", achieving a 98% accuracy rate.

• Research Assistant in Computer Vision - University Of Huddersfield (2024 - Present)

  • Leading research in advanced computer vision techniques, focusing on object detection, recognition, and deep learning methodologies.
  • Developing novel algorithms to enhance image analysis and tracking systems used in industrial applications, improving accuracy and efficiency.
  • Collaborating with interdisciplinary teams to integrate AI models into real-world scenarios and publishing findings in academic journals.

• Research Technician - University Of Huddersfield (2024 - 2024)

  • Led the development of a sophisticated multi-agent chatbot system utilizing LangChain, LangGraph, and the OpenAI API, optimizing customer interaction.
  • Implemented Retrieval-Augmented Generation (RAG) techniques, increasing chatbot response accuracy by 25%.
  • Worked with project engineers to compile FAQs, ensuring the chatbot handles 90% of customer inquiries effectively.

• Machine Learning Engineer - University Of Huddersfield (2023 - 2024)

  • Developed and implemented a deep learning model for object recognition and tracking in rail engines, achieving a 13% improvement in accuracy, elevating model performance from 80% to 93%.
  • Maintained and enhanced software for image analysis and detection of pallet racking, improving image-based task efficiency by 5%, resulting in a rise from 92% to 97% efficiency.

• Principal Software Engineer - Welltravel, United Kingdom (Remote, Part-time) (2022 - 2023)

  • Led the design and implementation of web applications using Ruby on Rails, significantly improving application performance by 40% through optimization of codebase, PostgreSQL queries, and caching mechanisms, thus supporting 3x more concurrent users.

• Chief Technology Officer - Welltravel, Bangladesh (2020 - 2022)

  • Directed a comprehensive technology roadmap, leading planning and execution across multiple teams to align technology initiatives with business goals, advancing the company's software solutions.
  • Spearheaded digital transformation initiatives, enhancing operational efficiencies and empowering teams with agile methodologies.

• Senior Software Engineer - Welltravel, Bangladesh (2017 - 2020)

  • Mentored 15 new engineers in Ruby on Rails development and agile project management, improving project delivery timelines by 25%.
  • Led the resolution of critical technological issues, elevating customer satisfaction and fostering a culture of continuous improvement.

• Languages: Ruby, Python, PHP, JavaScript, Java
• Frameworks: Ruby on Rails, TensorFlow, Keras, PyTorch
• Tools: Git, Docker, AWS, Heroku, PostgreSQL, MySQL
• Frontend: React
• Data Manipulation & Visualization: Pandas, Matplotlib

• Published Research Papers:
  • "Isolated Bangla Handwritten Character Recognition with Convolutional Neural Network" - Demonstrates my expertise in leveraging CNNs for recognizing handwritten Bangla characters, contributing significantly to automated document processing and linguistic studies. IEEE Link
  • "Boltvision: A Comparative Analysis of CNN, CCT, and ViT in Achieving High Accuracy for Missing Bolt Classification in Train Components" - This publication presents a groundbreaking comparative study highlighting my development of a model that accurately identifies missing bolts in train components, a critical aspect of railway safety. MDPI Link
  • "State-of-the-art Bangla Handwritten Character Recognition Using a Modified ResNet-34 Architecture" - In this work, I enhanced a ResNet-34 architecture to set a new standard for Bangla handwritten character recognition, showcasing my ability to improve upon existing deep-learning models for specific linguistic applications.
  • "Attention-Based Automated Pallet Racking Damage Detection" - This article introduces an attention-based mechanism for detecting damages in pallet racking systems, illustrating my innovative approach to ensuring warehouse safety through advanced image analysis techniques.
  • "YOLOV1 to YOLOV10: A Comprehensive Review of YOLO Variants and Their Application in the Agricultural Domain" - Provides an in-depth review of various YOLO models, their evolution, and their applications in agriculture.
  • "Comparative Analysis of YOLOv8 and YOLOv10 in Vehicle Detection: Performance Metrics and Model Efficacy" - This paper presents a detailed comparison of YOLOv8 and YOLOv10 models, focusing on their effectiveness in vehicle detection across various environments. The research contributes to the development of more efficient and accurate vehicle detection systems, particularly in intelligent transportation and autonomous driving. MDPI Link

With my background in software engineering and my growing expertise in artificial intelligence, my goal is to establish a career in the UK, focusing on computer vision and CNN. I am excited to apply my skills and knowledge to contribute meaningfully to the field of AI and develop innovative solutions to real-world problems.

Feel free to connect with me on LinkedIn or Twitter. I'm always open to discussing new opportunities, collaborating on projects, or just chatting about AI and technology!