Muhammad Majid Ahmad
BS Artificial Intelligence Graduate
I help businesses work smarter, faster, and better — using AI, automation, and data to create results you can actually measure
BS Artificial Intelligence Graduate
I help businesses work smarter, faster, and better — using AI, automation, and data to create results you can actually measure
Hi, I'm Majid Ahmad — I build machines that don't just think, they work for you. I'm an AI graduate who fell in love with making tech simple, transparent, and actually useful.
Over the past 3+ years, I've automated everything from tedious data scraping to full-blown business workflows, saving clients hours they didn't even realize they were wasting.
I mix machine learning, NLP, and custom automation to solve problems in real life, not just on paper. Whether it's a movie recommender, a credit-risk model, or a tool that pulls thousands of data points in minutes, I care about one thing: making ideas real, fast.
When I'm not coding, I'm teaching, mentoring, and sometimes just poking at new tech until it does something interesting.
Scripts Delivered
Happy Clients
Years Experience
Graduate
I'm driven by one question: what can AI really do beyond memorizing patterns? My work focuses on building systems that turn data into real-world impact.
AI learns from humans, and humans are biased—so the challenge is obvious. I explore ways to strip bias from text data while still preserving meaning. That means developing techniques for fair knowledge extraction, text mining, and building knowledge graphs that aren't skewed by the flaws in their source material.
Business data is messy, huge, and often ignored. I use machine learning and NLP to cut through the noise—predict trends, generate reports, and create AI systems that understand context, not just numbers. The goal: give decision-makers insights they can act on, not just dashboards to stare at.
Automation is great—until it breaks at the first curveball. I design AI-powered workflows that adapt in real time. From large-scale web scraping to smart monitoring tools, my systems learn from the process itself and adjust to keep things running without babysitting.
Academic journey and qualifications
Islamia University Bahawalpur
2021-2025 | Bahawalpur, Pakistan
CGPA: 3.58/4.0
Relevant Coursework: Machine Learning, Deep Learning, Data Structures, Algorithms, Computer Vision, Natural Language Processing
Government Sadiq Egerton College Bahawalpur
2019-2021 | Bahawalpur, Pakistan
Marks: 957/1100 (87%)
Subjects: Biology, Physics, Chemistry
Government Technical High School Bahawalpur
2017-2019 | Bahawalpur, Pakistan
Marks: 987/1100 (89.7%)
Subjects: Biology, Physics, Chemistry
Under Supervision of Sir Nasir, Additional Director IT, Islamia University Bahawalpur
2025
Research Focus: Conducted advanced research in Natural Language Processing (NLP), focusing on novel Relative Document Count (RDC)-based feature selection methods: BRDC, IRDC, and Dynamic RDC.
Methodology: Combined Positive Relevance Factor (PRF) and Negative Relevance Factor (NRF) into hybrid weighting schemes, outperforming traditional statistical measures like Chi-Square in both binary and multi-class text classification tasks.
Dataset Applications: Applied models to IMDB Movie Reviews (sentiment analysis) and 20 Newsgroups (topic classification) datasets, achieving higher accuracy and better generalization through optimized feature selection and Support Vector Machine (SVM) classifiers.
Algorithm Innovation: Engineered dynamic, noise-resistant algorithms capable of adapting feature selection strategies to dataset characteristics, improving robustness and efficiency in high-dimensional text mining problems.
Research Pipeline: Designed a fully reproducible research pipeline with:
Research Impact: Research outcomes contribute to the academic advancement of text classification by providing scalable, dataset-adaptive methods applicable in sentiment analysis, topic detection, and automated content filtering.
A showcase of academic innovations and freelance solutions across diverse domains.
Platform: Independent Research | Islamia University Bahawalpur | 2024–2025 | Bahawalpur, Pakistan
Researched and developed a computer vision–based system that extracts structured data from YouTube videos without relying on APIs or transcripts. Combined EasyOCR, OpenCV, and PyTorch for text recognition, built a screenshot-based ROI annotation tool for precise zone selection, and implemented multi-threaded processing for large datasets. Delivered a cost-effective alternative to commercial solutions, setting new benchmarks for accuracy and efficiency in visual data extraction.
Key Contributions:
Technologies: Python, BeautifulSoup, Selenium, Requests
Platform: Client Project | 2024–2025
Built a Python-based scraper to pull detailed product data—titles, prices, ratings, reviews, and specs—from specific Amazon categories without using the Amazon API. Data is stored in CSV/JSON for analysis and inventory management.
Technologies: Python, Selenium, Hidden APIs, Pandas
Platform: Client Project | 2024–2025
Created a script to capture numerical values from online bar graphs and charts. Used Selenium automation to navigate interfaces and tapped into undocumented APIs to access raw data, exporting results into CSV for analysis.
Technologies: Python, Selenium, BeautifulSoup, Pandas, Excel Processing
Platform: Client Project | 2024–2025
Developed an Instagram analytics tool to scrape profile data (followers, following lists) and reel metrics (likes, views, comments, hashtags). Built an automated pipeline to process multiple profiles and compile engagement stats into structured Excel reports.
Technologies: Python, Selenium, BeautifulSoup, Pandas, Excel Processing
Platform: Client Project | 2024–2025
Built an automated TikTok scraper that processes sound links from Excel, finds all profiles using those sounds, and extracts contact details (emails, social links, phone numbers). Designed as a full end-to-end pipeline, it handles multiple sound links in bulk and compiles all results into structured Excel reports.
Technologies: Python, Selenium, BeautifulSoup, Google Maps API
Platform: Client Project | 2024–2025
Created a data extraction system to compile a complete list of New York City convenience stores from Google Maps and Yellow Pages. Collected over 1,000 entries with business names, addresses, contact details, and operational information, stored in a structured database for easy querying.
Technologies: Python, Selenium, Excel Processing, Residential Rotating Proxies, 2captcha API, Anti-Bot Detection
Platform: Client Project | 2024–2025
Automated the Portuguese visa application process by reading applicant data from Excel, uploading required documents, and completing complex forms. Integrated residential rotating proxies, 2captcha API for CAPTCHA solving, and advanced anti-bot detection to ensure smooth, large-scale processing of multiple applications.
Technologies: Python, BeautifulSoup, Selenium, Email Automation, Database
Platform: Client Project | 2024–2025
Developed a system to monitor product prices across multiple e-commerce sites, track historical trends, and detect changes. Sends automated email and SMS alerts based on user-defined thresholds, with full historical data storage for trend analysis.
Technologies: Python, Selenium, Streamlit, Excel Processing, Keyword Detection
Platform: Client Project | 2024–2025
Built a Streamlit-based Facebook automation tool that allows users to input a specific keyword. The system launches Facebook, sets filters like date and location, and automatically scrolls through posts. It scans post content for the given keyword, and when a match is found, it posts a predefined comment. For each matched post, it extracts and saves the user’s name, profile link, post content, and matched keyword into an Excel file.
Technologies: Python, Selenium, Excel Processing, Session Management, Cookie Handling
Platform: Client Project | 2024–2025
Created an automation tool that reads contacts from Excel and messages from text files to send personalized WhatsApp messages and media in bulk. Maintains authentication using advanced session and cookie management for uninterrupted operation.
Technologies: Python, Pandas, NumPy, Scikit-learn, NLTK, Cosine Similarity, Feature Engineering
Platform: Academic Project | 2024–2025
Built a content-based movie recommendation system using the TMDB 5000 movies dataset. Processed and cleaned metadata (genres, cast, crew, plot summaries), extracted features with TF-IDF vectorization, and applied cosine similarity to recommend movies based on content relevance. Delivered personalized suggestions with similarity scores and detailed movie info.
Technologies: Python, Scikit-learn, XGBoost, SMOTE, Pandas, NumPy, StandardScaler
Platform: Academic Project | 2024–2025
Designed a credit risk prediction model to forecast loan defaults for financial institutions. Preprocessed data using median imputation, one-hot encoding, and normalization via StandardScaler. Balanced datasets with SMOTE and tested multiple algorithms including Logistic Regression, Random Forest, and XGBoost—achieving 94% accuracy with Random Forest, outperforming others in precision and recall.
Technologies: Python, Scikit-learn, NLTK, Pandas, Matplotlib, TF-IDF, SMOTE
Platform: Academic Project | 2024–2025
Developed a sentiment analysis system to classify text data into positive, negative, or neutral categories for feedback, reviews, and social media comments. Implemented tokenization, stopword removal, stemming, and TF-IDF vectorization for feature extraction. Balanced the dataset with SMOTE and tested Logistic Regression, SVM, and Naive Bayes—achieving 88% accuracy with Logistic Regression as the top performer.
Technologies: Python, Scikit-learn, NLTK, Pandas, Streamlit, TF-IDF, Grid Search
Platform: Academic Project | 2024–2025
Built an interactive text classification app to categorize input text into Business, Health, or Sports. Used lemmatization, bigram TF-IDF features, and a Streamlit interface for real-time predictions. Applied Grid Search for hyperparameter tuning, achieving 90% accuracy with Logistic Regression.
Technologies: Python, Django, Streamlit, yt-dlp, HTML/CSS, JavaScript
Platform: Personal Project | 2024–2025
Built a user-friendly web app for downloading YouTube videos and audio in multiple formats (MP4, MP3, various quality levels). Implemented URL validation, download progress tracking, and a file management system. Used yt-dlp for reliable media extraction and conversion, with Django and Streamlit handling backend and interface.
Technologies: Python, Streamlit, Pandas, Plotly, Matplotlib, Seaborn, NumPy
Platform: Client Project | 2024–2025
Created a Streamlit-based data analysis platform allowing clients to upload Excel files and get instant insights. Automated statistical summaries, interactive charts, trend detection, and report generation. Integrated Plotly and Matplotlib/Seaborn for rich, interactive visualizations with downloadable reports for business decision-making.
Technologies: Python, Streamlit, Pandas, NumPy, Plotly, Matplotlib, Seaborn
Platform: Client Project | 2024–2025
Built a healthcare analytics dashboard to assess doctor performance across metrics like patient volume, revenue generation, and tenure. Conducted correlation analysis to uncover performance patterns, implemented ranking systems for top and bottom performers, and created interactive visualizations for trend analysis. Delivered actionable insights to healthcare managers through statistical reporting and dynamic charts.
Continuous learning and professional development
Comprehensive AI course covering search algorithms, knowledge representation, uncertainty, optimization, machine learning, neural networks, and large language models.
Comprehensive PyTorch fundamentals covering tensor operations, data types, and deep learning foundations. Mastered tensor creation, manipulation, and GPU computing basics.
Learned and implemented FastAPI to build high-performance RESTful APIs using Python. Covered topics include routing, dependency injection, Pydantic schemas, database integration (SQLAlchemy), and async endpoints. Gained hands-on experience by building real-world projects with proper folder structure and modularization.
Advanced Machine Learning, Artificial Intelligence, and Data Science comprehensive certification
Advanced machine learning algorithms, data science techniques, and model deployment
Professional web scraping techniques using Python, BeautifulSoup, and Selenium
Comprehensive Python and Django web development course covering fundamentals to advanced concepts
Web application development using Python and Flask framework with database integration
Expertise across multiple domains
Advanced Python development and automation
OOP, Data Structures, Algorithms, Libraries, Frameworks, Testing
Advanced web scraping and process automation
Selenium, BeautifulSoup, Scrapy, Requests, Multi-threading, Anti-detection
ML algorithms and deep learning frameworks
Scikit-learn, TensorFlow, PyTorch, Keras, Computer Vision, Model Deployment
Statistical analysis and data visualization
Pandas, NumPy, Matplotlib, Seaborn, Plotly, Statistical Modeling
Advanced text intelligence and language understanding
NLTK, spaCy, Transformers, BERT, Sentiment Analysis, Named Entity Recognition
Full-stack web application development
HTML, CSS, JavaScript, Bootstrap, Flask, Django, Streamlit
Database design and management
MySQL, PostgreSQL, MongoDB, SQLite, Database Design
Recognitions and achievements in freelancing and professional excellence
Platform: Upwork | 2025
Earned Top Rated status as a Python Developer, recognized for delivering high-quality projects, maintaining exceptional client satisfaction, and demonstrating consistent performance on the platform.
Platform: Fiverr | 2023
Achieved Level 2 Seller status specializing in Web Scraping and Automation, backed by consistent 5-star reviews and a high order completion rate.
Organization: NAVTTC | 2025
Served as a certified trainer for the Advan Honored as a certified trainer for the Advanced Python Programming Course at DevCastle, IUB under NAVTTC. Delivered structured training sessions focusing on modern Python practices.
Let's discuss research opportunities and collaborations
Let's connect on WeChat.
ID: heresmajid
based in
Bahawalpur, Pakistan
Connect with me on professional platforms