Projects using PyTorch in LahoreProjects using PyTorch in Lahore
Cover image for Fine-tuned OpenAI Whisper model on
Fine-tuned OpenAI Whisper model on domain-specific medical audio data to improve transcription accuracy for clinical and healthcare use cases. The project involved preprocessing medical speech datasets, handling noise and terminology challenges, and optimizing the model for improved recognition of medical vocabulary, accents, and context-heavy conversations. Delivered a robust speech-to-text system capable of producing highly accurate, structured transcriptions suitable for documentation, reporting, and downstream healthcare applications.
0
143
Cover image for AI Vision for Retail, Industrial
AI Vision for Retail, Industrial & Monitoring Workflows Overview I have built and deployed multiple real-world computer vision systems for industrial inspection, retail automation, and monitoring workflows. My responsibilities covered: 🔹 Dataset preparation and labeling 🔹 Object detection model training 🔹 Segmentation model training 🔹 YOLO-based detection and tracking 🔹 Image/video inference pipeline development 🔹 Model evaluation and threshold tuning 🔹 Production deployment support 🔹 Cloud server management and optimization 🔹 Building practical AI workflows for real-world operational environments Fish Quality Inspection System - lythium.cl (http://lythium.cl) I led the development of an advanced fish quality inspection solution for an industrial workflow. The system used image analysis to monitor fish quality and support automated fish sorting based on AI predictions. 🔹 Led the development of an advanced AI-powered fish quality inspection system for an industrial workflow. 🔹 Built an image analysis pipeline to monitor fish quality from production-line images. 🔹 Trained object detection models to identify fish and relevant visual quality indicators. 🔹 Trained segmentation models to support more detailed visual inspection of fish regions. 🔹 Designed the AI workflow to support automated fish sorting based on model predictions. 🔹 Worked on inspection logic that could classify or route fish based on quality-related outputs. 🔹 Designed the system for conveyor-belt usage, where images need to be processed consistently and reliably. 🔹 Focused on production issues such as image quality, camera consistency, lighting variation, and model reliability. 🔹 Helped convert visual inspection from a manual/rule-based workflow into an AI-supported inspection pipeline. 🔹 Built the system to reduce manual inspection effort and improve production workflow efficiency. Shelfr.ai (http://Shelfr.ai) - Retail Automation Platform I developed AI image solutions for retail automation and execution. The system handled large-scale product detection across 10,575+ SKUs, price tag detection, shelf and display type detection, and gap detection for empty shelf spaces. 🔹 Developed large-scale AI image solutions for retail automation and execution. 🔹 Worked on product detection across 10,575+ SKUs, where each SKU represented a unique product. 🔹 Built object detection workflows to identify products from retail shelf images. 🔹 Developed price tag detection to locate and extract price label areas from store images. 🔹 Worked on shelf and display type detection to understand the retail environment layout. 🔹 Built gap detection logic to identify empty shelf spaces and out-of-stock areas. 🔹 Supported computer vision workflows for retail compliance, shelf monitoring, and store execution. 🔹 Worked with high-volume image data and production-level inference requirements. 🔹 Managed high-load production servers on Google Cloud Platform. 🔹 Implemented load balancing and autoscaling to improve system stability under production traffic. 🔹 Focused on scalable AI infrastructure capable of handling real-world retail image workloads. 🔹 Helped create AI systems for inventory visibility, shelf condition monitoring, and retail execution analytics. lake-shield.com (http://lake-shield.com) - USA LAKES - Boat Detection & Inspection System 🔹 Worked on a YOLO-based boat detection, tracking, and monitoring system. 🔹 Labeled datasets for boat detection and inspection model training. 🔹 Prepared image/video data for object detection training workflows. 🔹 Trained YOLO object detection models to detect boats in monitoring footage. 🔹 Built a detection pipeline capable of identifying boats from visual data. 🔹 Worked on boat tracking logic to monitor boat movement across frames. 🔹 Supported inspection and monitoring workflows using computer vision predictions. 🔹 Developed an end-to-end pipeline from labeled data to trained model and inference output. 🔹 Focused on practical model performance in outdoor environments where lighting, distance, angle, and background can vary. 🔹 Helped build a monitoring system that could support automated detection and review instead of fully manual observation. My Responsibilities Across These Projects 🔹 Led AI/computer vision system development 🔹 Designed labeling and dataset preparation workflows 🔹 Trained YOLO/object detection models 🔹 Trained segmentation models where needed 🔹 Built image and video inference pipelines 🔹 Evaluated models using practical production metrics 🔹 Improved model performance through dataset cleanup, retraining, and threshold tuning 🔹 Integrated AI models into backend or operational workflows 🔹 Supported production deployment and infrastructure optimization 🔹 Worked with real-world constraints such as lighting, camera angle, image quality, latency, and false detection rates Technologies Used 🔹 Python 🔹 YOLO / YOLOv8 🔹 Object Detection 🔹 Image Segmentation 🔹 OpenCV 🔹 PyTorch 🔹 FastAPI 🔹 Google Cloud Platform 🔹 Linux Servers 🔹 Load Balancing 🔹 Autoscaling 🔹 Custom Data Labeling Workflows 🔹 Model Training 🔹 Model Evaluation 🔹 Inference Pipeline Development 🔹 Production AI Deployment
1
123
Cover image for A dark-studio, sketchbook-themed portfolio for
A dark-studio, sketchbook-themed portfolio for a portrait/line artist. Same stack as the AI-engineer portfolio: Next.js (App Router), Tailwind, Framer Motion, lucide-react, react-icons, and a working nodemailer contact form. The big thing is the AI , image to 3d model generator, which uses a tencent cloud API Stable Diffusion model to generate 3D models from 2D images. The models are then exported to GLTF format and displayed in the portfolio using Three.js.
0
24
Cover image for These are 3 projects i
These are 3 projects i did as a solo dev
0
30
Cover image for Mental health multi ai agent
Mental health multi ai agent platform
0
27
Cover image for Urdu News AI Summarization &
Urdu News AI Summarization & Search System Designed and developed an end-to-end AI platform for processing Urdu news from television broadcasts and digital media. The system automatically transcribes speech, extracts text from news tickers, generates concise summaries, and enables intelligent semantic search across large collections of Urdu news content.
0
30
Cover image for An AI-powered fake medicine detection
An AI-powered fake medicine detection system built with Python, PyTorch, NumPy, and OpenCV. The system processes medicine package images, extracts medicine names using OCR, and verifies them against a dataset of registered medicines to help classify products as real or fake.
1
36
Cover image for Beyond the Naked Eye — Automated Delivery Analysis in Cricket
Beyond the Naked Eye — Automated Delivery Analysis in Cricket
0
72
Cover image for A master's thesis titled "Global
A master's thesis titled "Global Optimization for PV-Integrated Smart Distribution Grids Using Data-Driven Neural Networks," supervised by Dr. Raheel Zafar and evaluated by Dr. Nauman Zafar Butt. The core idea is a teacher-student setup: an SOCP relaxation acts as a convex "teacher" that produces near-optimal solutions to a voltage-var optimization (VVO) problem, and a feedforward neural network (FCNN) is trained as a fast "student" surrogate that mimics the teacher at inference speed. A constraint-embedding trick using a tanh-bounded output layer cut infeasibility rates from 76.61% down to 0.13%. Validated on the IEEE 33-bus system using OpenDSS, with cyclical time encoding to capture daily/seasonal patterns in the grid data.
0
20