Syed (Sabbi ui) H.
AI Intern / Front End Developer Intern
Experience
AI Intern / Front End Developer Intern
Swati Tech
- Worked on audio and image classification models using TensorFlow, Scikit-learn, and Keras
- Performed model training, hyperparameter tuning, and evaluation in Jupyter Notebook and Google Colab
- Applied data preprocessing, augmentation, and feature extraction techniques for improved accuracy
- Contributed to research-oriented AI tasks, including real-time model deployment
- Collaborated with mentors on AI automation workflows and edge AI experiments involving audio event detection and computer vision
- Developed responsive web interfaces using React.js, HTML, CSS, and JavaScript
- Integrated REST APIs for dynamic data display and maintained clean, reusable components
- Used Git for version control and participated in code reviews to ensure consistency and maintainability
Final Year Project - Illegal Wood Cutting Detection
- Developed a real-time audio classification system in Python using TensorFlow, YAMNet, RandomForest, TensorFlow Lite, Arduino IDE, and C++ to detect illegal wood cutting sounds in forest environments
- Trained a YAMNet-based transfer learning model on a custom dataset with classes like Chainsaw, Axe, Handsaw, Bird Chirping, Generator, Thunderstorm, and Fire
- Applied data augmentation techniques including white noise injection, pitch shifting, and time-stretching to enhance model robustness
- Achieved 87% accuracy with class balancing, stratified splits, and class weighting
- Converted and quantized the model to TFLite for deployment on Arduino Nano 33 BLE and Portenta H7 using real-time mic input and log-mel spectrogram preprocessing
Personal Project - AI-Powered Automatic Sky Replacement
- Built an AI-based image enhancement pipeline in Python using PyTorch, OpenCV, SkySegNet, and SkyAR to automatically replace skies in real-world images
- Developed a fully automated workflow covering sky detection, segmentation, resizing, replacement, and seamless blending
- Achieved 90% segmentation accuracy across varied lighting and weather conditions
- Reduced manual intervention to zero
Personal Project - Text-to-Speech
- Developed and automated an AI text-to-speech workflow in n8n using OpenAI APIs to generate realistic human voices from text
- Built a fast, scalable automation pipeline for content creation, accessibility, and voice assistant use cases
Personal Project - University Merit Prediction System
- Built an ML web app in Python using Scikit-learn and Gradio to predict NUST departments and schools from NET merit positions using 2023-2024 data
- Achieved 89% accuracy
- Deployed a real-time interactive interface on Google Colab and hosted it on Render with a custom purchased domain
General Secretary
Ushers and Event Management Society
- Organized and led large-scale cultural and entertainment events with over 2,000 attendees at the university
- Managed end-to-end event operations, including ticketing, artist coordination, stage lighting, sound systems, decorations, and hosting
- Supervised a team of volunteers to ensure smooth execution and exceptional audience experience
- Successfully delivered two major annual events (Mega EVE and Mega EVE 2.0) in 2022 and 2023 with strong organizational and leadership outcomes
Summary
Al/ML Engineer, Deep Learning, Computer Vision, Audio Intelligence
Experienced in model deployment pipelines, API integration, and workflow automation using n8n and LLMs for scalable AI systems. Results-driven Computer Science graduate with hands-on experience in machine learning, deep learning, and AI automation. Skilled in developing and deploying end-to-end intelligent systems from audio classification (TensorFlow, YAMNet, Arduino) to computer vision (PyTorch, SkySegNet, OpenCV) and automation workflows (n8n, OpenAI, RAG, MCP, Agents). Proven ability to build real-time, production-ready ML apps using Python, Scikit-learn, TensorFlow Lite, and Gradio, with deployments on Render and microcontroller hardware. Passionate about building efficient, real-world AI systems that combine accuracy, automation, and scalability.
Languages
Education
Pak Austria Fachhochschule Institute of Applied Science and Technology
Bachelor in Computer Science · Computer Science · Pakistan
Similar Freelancers
Discover other experts with similar qualifications and experience
- English
- Deutsch
2025 © FRATCH.IO GmbH. All rights reserved.