Shakeel Ur Rehman's Work | ContraWork by Shakeel Ur Rehman
Shakeel Ur Rehman

Shakeel Ur Rehman

AI/ML Engineer | Computer Vision | Agentic AI | OCR | YOLO

    3
    Followers
Followed by Lilya Y, Shanmukhi B, and GALLERY L

Faisalabad, Pakistan

WorkPostsServicesAbout

Faisalabad, Pakistan

Anti Drone Telemetry Defense System Engineered an autonomous, military-grade Counter-UAS (Unmanned Aerial System) intelligence and early warning defense system for perimeter security, airspace monitoring, and critical infrastructure defense. It detects micro/mini drones, tracks them across dynamic video feeds, and computes real-time 3D spatial telemetry vectors and geodetic GPS coordinates. Key Architectural Highlights: High-Resolution Sliced Detection (SAHI + YOLO11) Integrated Slicing Aided Hyper Inference (SAHI) with an optimized YOLO11 ONNX pipeline to eliminate small-target loss, detecting distant, low-RCS micro-drones at up to 5x the range of conventional downsampled models. Robust Multi-Object Tracking & Threat Gate DeepSORT with Cosine appearance metric plus a temporal state-machine (8-consecutive-frame confirmation) suppresses false positives from birds, insects, and atmospheric clutter before a target moves from ACQUIRING to LOCKED THREAT. Real-Time Spatial Telemetry & Geodesy Engine True compass bearing (0°–360°) and 16-point cardinal direction (N, NE, E, etc.), synced with camera boresight azimuth and elevation Triangle-similarity optical slant and ground distance estimation Drone GPS projection (Latitude/Longitude) via Vincenty's Direct Geodesic Algorithm on the WGS-84 ellipsoid C2 Dashboard & Autonomous Alerting Live tactical web dashboard with CRT radar sweep and telemetry streamed via Server-Sent Events (SSE) Multi-channel threat dispatch: RF Comset simulation, local sirens, audit logging Forensic evidence vault: watermarked 5-second MP4 incident clip generated on every threat lock Tech Stack: Python, YOLO11, SAHI, DeepSORT, OpenCV, ONNX Runtime, Flask, WGS-84 Geodesy, NumPy, Multi-Threading, Real-Time Edge Vision
0
47
End-to-End Telco Customer Churn Prediction System Developed and deployed a production-grade Machine Learning web application designed to predict telecom customer churn and identify key drivers of customer attrition. This project bridges the gap between data science research and real-world deployment by transforming a Jupyter Notebook workflow into an interactive, cloud-hosted web service. Key Technical Highlights & Workflow: Exploratory Data Analysis & Preprocessing: Conducted thorough EDA on customer behavioral data, handled missing values, encoded categorical features, and performed feature scaling to ensure robust model performance. Model Training & Hyperparameter Tuning: Trained multiple machine learning algorithms (including Random Forest, XGBoost, and Logistic Regression) and optimized hyperparameters to achieve high predictive accuracy and generalization. Model Explainability (SHAP): Integrated SHAP (SHapley Additive exPlanations) values to make complex model predictions transparent, helping stakeholders easily interpret feature impacts (such as tenure, monthly charges, and contract types) on individual predictions. Backend & API Architecture: Built a high-performance RESTful API using FastAPI to handle real-time inference requests seamlessly, coupled with a clean, responsive frontend user interface. Cloud Deployment & DevOps: Successfully packaged the application and trained model (.pkl), navigated serverless deployment constraints by leveraging container-based cloud infrastructure (Render), and ensured continuous availability. Tech Stack: Python, Scikit-Learn, XGBoost, Pandas, NumPy, FastAPI, Uvicorn, SHAP, HTML/CSS, Git, Render.
0
50
App Name: GigPulse Category: 📊 Dashboard + 🤝 Social/Community What it does: GigPulse is a personal freelance project tracker built for independent professionals who juggle multiple platforms. Freelancers can log their projects, track earnings per platform, monitor project status, and attach video demos of their work all in one private dashboard. The community benchmarking layer lets users see anonymized stats from other freelancers in their niche, so you always know how you stack up. How I built on top of the AI-generated base: Bubble AI generated the core dashboard structure and data architecture. From there, I manually engineered the video upload and playback system per project, built privacy-filtered data workflows so each user only sees their own data, added platform-specific earnings tracking with status badges, and designed a clean dark UI with collapsible video players. The AI gave me the skeleton, I built the muscles. This is the Bubble AI-generated base after my initial prompt 'Build a freelance analytics dashboard app called "GigPulse" where freelancers can track their gigs, proposals sent, earnings per platform, client response rates, and monthly revenue goals. Include a community benchmarking section where users can see anonymized stats from other freelancers in their niche. Add user authentication, a personal stats dashboard with charts, a goal-setting tracker, and a community feed where freelancers share weekly wins.' Everything you see in the final build beyond this point was manually engineered.
1
4
403
Sign Language Detection
0
103