Hello, I'm
Abdallah Waleed
Data Scientist & IoT Engineer
Passionate about leveraging data analytics and IoT technologies to build intelligent, scalable systems
Hello, I'm
Passionate about leveraging data analytics and IoT technologies to build intelligent, scalable systems
I am a passionate Data Scientist, Data Analyst, and IoT Engineer. I specialize in bridging the gap between physical hardware and intelligent data systems, transforming real-world sensor streams and IoT telemetry into actionable insights and predictive models. My expertise spans the entire lifecycle of data and hardware integration: from prototyping smart devices with Arduino & ESP32 to deploying machine learning models, conducting advanced statistical analysis, and designing interactive data dashboards. By merging edge electronics with scientific data modeling, I architect end-to-end smart solutions that drive data-driven decision-making.
Years Experience
IoT Projects
Data Projects
AI Projects
End-to-end IoT solution architecture including hardware selection, sensor integration, and connectivity design
Transforming raw data into actionable insights through dynamic dashboards, data modeling, ETL cleaning, and executive reporting using Power BI, Tableau, and Excel.
Building predictive models, statistical forecasting pipelines, and advanced NLP systems using Python, Machine Learning regression/classification, and automated tracking.
Improved command response by 35%, reaching <500 ms latency and 95% control accuracy. Enabled stable wireless communication within 10m range using Bluetooth (HC-05). Enhanced object detection by 100% coverage in 10-150cm range through sensor calibration.
Built an intelligent voice-controlled car system integrating AI, IoT, and mobile technologies. Trained an NLP Intent Classification model (Logistic Regression + TF-IDF) achieving ~93% accuracy for Arabic and English voice commands. Developed a Flask REST API for real-time command processing and deployed it using ngrok. Implemented secure IoT communication using ESP8266 with MQTT over TLS (HiveMQ Cloud) for real-time motor control via Arduino and L298N driver. Created a Flutter mobile app with speech-to-text, Arabic language support, and persistent settings. The complete system flow: Voice Command → Flutter App → Flask API → NLP Model → MQTT → ESP8266 → Motors, demonstrating end-to-end integration of AI, cloud services, and embedded systems.
Designed an interactive Power BI dashboard analyzing sales performance across regions, product categories, and time periods. Built calculated KPIs and DAX measures for revenue tracking, growth trend analysis, and top-performer identification, enabling executive-level reporting.
Built a comprehensive Power BI dashboard for a retail food mart chain, visualizing inventory movement, sales volume, and outlet level performance. Applied data cleaning and transformation techniques to ensure accuracy across multiple data sources, delivering clear operational insights to support inventory and sales strategy decisions.
Built a comprehensive interactive sales analysis dashboard for a B2B Reseller-Based Sales System. Covers total sales analysis, profit & cost tracking, sales trends over time, product performance, profit by category, and reseller insights with interactive filters & KPIs.
Developed a comprehensive Islamic companion app featuring a complete Quran reader, accurate GPS-based prayer times, Azan notifications, and daily Hadith alerts. Implemented a smart notification system delivering Azkar, Salawat, and Tasbih reminders every 10 minutes around the clock, with dedicated Ramadan mode and Iftar dua alerts.
Built a comprehensive Excel control dashboard to track a department's performance across a full month, integrating 8 operational areas (security incidents, patrols, fleet status, workforce readiness, etc.). Features raw data and display sheets for each module, and a central KPIs dashboard view.
Designed and developed an interactive Sales Performance Dashboard for a virtual electronics store called “TechVision” using Microsoft Power BI. Built a complete Data Model with relationships between Sales and Products tables. Created dynamic DAX measures for Total Revenue, Cost, and Profit. Designed interactive visualizations including a Revenue Trend Line, Profit by Country Bar Chart, and Revenue by Category Donut Chart. Added interactive slicers and customized UI/UX.
Built a complete end-to-end regression pipeline on a real Egyptian real estate dataset (1,008 listings). 11-stage pipeline: EDA, missing value imputation, outlier removal (IQR), feature scaling, and 6 trained models. Lasso achieved R²=0.9730 with MAE ≈ 24,000 EGP — L1 automatically zeroed weak features for built-in feature selection. All models tracked with MLflow. Includes an interactive browser-based web app with real-time prediction sliders.
Have feedback? Share your experience!
Submit Feedback+20 100 58 31 772
waleedabdallah238@gmail.com
Your feedback helps me improve and showcase my work better. Share your experience working with me or your thoughts on my portfolio.