AI Developer | Built 15+ AI Agents

github

Omar Nahdi • Peerlist

RustSight is an open-source command-line tool designed for fast CSV dataset profiling and validation, making it an essential asset for data engineers and AI/ML practitioners. Built with Rust, this tool excels in speed and efficiency, outperforming traditional tools like Pandas by 6.1x on large datasets. RustSight is capable of detecting missing values, outliers, type mismatches, and no-variance columns, ensuring data quality before AI/ML model training. Key features include:<ul><li>CSV dataset profiling with column type detection</li><li>Missing value analysis per column</li><li>ML readiness validation with outlier detection</li><li>Streaming processing for large files</li><li>Automatic report generation</li></ul>RustSight supports CSV and TXT/Binary formats, with plans to include Parquet, JSON, and Arrow. The tool's performance is benchmarked against industry standards, demonstrating its capability to handle large datasets efficiently. Installation is straightforward via Cargo, requiring the Rust toolchain. Users can quickly profile datasets and validate them for ML readiness, making RustSight a powerful tool for data validation and profiling tasks.

RustSight Platform

RustSight is a fast, safe, and extensible dataset analysis CLI tool written in Rust. This tool is designed to perform data validation and exploratory analysis, crucial steps before AI/ML model training. It efficiently handles any CSV file and can analyze binary or text files to extract useful properties.<h3>Key Features</h3><ul><li>CSV Dataset Analysis: Detects numeric vs categorical columns, counts missing values per column, computes basic statistics (min, max, mean) for numeric columns, handles large CSV files efficiently through streaming, and generates a clean, readable analysis report.</li><li>Data Validation: Detects columns with high missing value ratios, flags no-variance columns, detects potential outliers, identifies mixed-type columns, and prints clear validation warnings before ML usage.</li><li>File Analysis: Counts total bytes, detects UTF-8 validity, counts lines and words for text files, and counts non-ASCII bytes for binaries.</li></ul>The tool is open source, licensed under the MIT License, and contributions are welcome. It is designed with a CLI-first approach, making it ideal for automation and scripting. The tech stack includes Rust for performance and memory safety, and the csv crate for efficient CSV parsing. Portfolio: <a target="_blank" rel="noopener noreferrer" class="break-all link-break text-gray-500 hover:underline" href="https://omarnahdi.dev">https://omarnahdi.dev</a>

RustSight

<h3>Narrative AI: Intelligent LinkedIn Post Generator</h3>Short DescriptionAn advanced AI-powered LinkedIn post generation system that creates engaging, research-backed professional content using intelligent agents and a beautiful modern interface.Detailed DescriptionNarrative AI revolutionizes LinkedIn content creation by combining the power of AI agents with comprehensive research capabilities. This open-source project leverages Google's Gemini AI and the Agno framework to deliver a sophisticated content generation experience that goes beyond simple text generation.Key Capabilities:<ul><li>Research-Driven Content: Automatically researches topics using Brave Search and web scraping to ensure posts are factually accurate and current</li><li>Intelligent Agent System: Utilizes multi-agent architecture for specialized tasks including research, content creation, and formatting</li><li>Beautiful Modern Interface: Features a stunning Agent UI built with Next.js that provides conversational interaction with persistent memory</li><li>Professional Formatting: Generates LinkedIn-optimized posts with proper hashtags, engagement hooks, and professional tone</li><li>Conversation Memory: Maintains context across interactions for refined and personalized content generation</li></ul>Perfect For:<ul><li>Content creators and social media managers</li><li>Professionals looking to maintain consistent LinkedIn presence</li><li>Marketing teams needing scalable content solutions</li><li>Anyone wanting data-backed, engaging professional posts</li></ul>Technical Highlights:<ul><li>Built with Python 3.11+ and modern frameworks</li><li>Powered by Google Gemini AI for advanced language understanding</li><li>Integrates Brave Search API for real-time research capabilities</li><li>Uses Playwright for advanced web scraping</li><li>SQLite-based agent memory for persistent conversations</li><li>Fully open-source with Apache License</li></ul>The system transforms simple prompts into comprehensive, engaging LinkedIn posts by researching the topic, analyzing the target audience, and crafting content that drives professional engagement. Whether you're discussing emerging technologies, industry trends, or thought leadership topics, Narrative AI ensures your content is both informative and compelling.Repository Stats<ul><li>Language: Python</li><li>License: Apache License</li><li>Framework: Agno AI Framework</li><li>Interface: Next.js Agent UI</li><li>Database: SQLite</li><li>AI Model: Google Gemini Pro</li></ul>

Narrative AI

Gemini Pro Vision ATS System stands out as a beacon of innovation in the realm of job application solutions. By seamlessly integrating cutting-edge technology, this project not only streamlines the application process but also offers a personalized touch through insightful feedback. Its scope extends beyond traditional ATS capabilities, ensuring that candidates receive tailored recommendations to enhance their resumes. Embrace the future of job applications with Gemini Pro Vision, where precision meets personalization to unlock new opportunities. 🚀💼

Advanced ATS system using Gemini Pro Vision -

Project Goal: This project aims to segment customers into distinct groups based on their characteristics and behavior patterns using K-Means clustering. This segmentation will help us better understand our customer base, personalize marketing efforts, and improve customer engagement and loyalty.

Customer Segmentation

Embarking on a journey at the intersection of technology and healthcare, I've developed a machine learning model for predicting diabetes risk. Through meticulous curation of diverse datasets, this project empowers healthcare professionals and individuals with an accurate and reliable tool for early detection. Join me in making a positive impact on diabetes diagnosis and management.

Diabetes Prediction Model

🍄 Mushroom Classifier Project 🌿 Dive into the fascinating world of mushrooms with my latest project! 🚀 Using Support Vector Machines (SVM) in machine learning, I've crafted a powerful model that predicts whether a mushroom is edible or poisonous based on its unique features. 📊 Key Highlights: 1. Machine Learning Magic: Witness how SVM transforms raw mushroom data into predictive insights. 2. Real-World Application: Explore the practical side as I demystify the process of predicting edibility using Python. 3. Code Sharing: Download the PDF to peek behind the curtain and uncover the secrets of the Mushroom Classifier. 🌐 Let's Learn Together: Join me on this journey of exploration and discovery! Feel free to dive into the code, ask questions, and let's cultivate our understanding of machine learning. 🌱💻

Prediction of Edible and Poisonous Mushrooms

Empowering Investment Decisions with Stock Price Prediction 📈💰 Developed a simple yet effective tool for predicting stock prices using linear regression and 50-day moving average (50DMA) data. 📊 Key Features: 1. Train a linear regression model using 50DMA data 2. Generate closing price predictions based on new 50DMA values 3. Visualize the relationship between 50DMA values and predicted closing prices Contributions: 1. Made the code available on GitHub for easy access and learning 2. Encouraged contributions from developers and data enthusiasts to enhance the project Impact: 1. Empowered individuals to make informed investment decisions 2. Provided valuable insights into stock price movements 3. Contributed to the advancement of stock prediction techniques

Prediction of Google's Stock Prices

🔍 Exploring the realm of cybersecurity and communication integrity with my latest project: 'Spam Detection with Naive Bayes Algorithm.' This endeavor showcases my commitment to leveraging cutting-edge machine learning for real-time identification and filtration of spam messages. Excited to contribute to a safer and cleaner digital communication landscape! 🚀🤖📊.

Spam Detection Model

🚀 Introducing SpamShield: A Machine Learning Model for SMS Spam Detection 🚀 SpamShield is a machine learning-powered tool designed to classify SMS messages as either spam or not spam with high accuracy. This project was created to explore and implement natural language processing (NLP) techniques and machine learning algorithms focused on text classification. Built with a user-friendly interface using Streamlit, SpamShield makes spam detection both simple and efficient. Key Features: 1. Real-Time SMS Classification: Instantly analyzes SMS text and flags potential spam. 2. NLP-Powered Filtering: Uses tokenization, stemming, and stop-word removal to process messages accurately. 3. User-Friendly Interface: Allows anyone to enter a message and receive immediate feedback. SpamShield is an ideal demonstration of how machine learning and NLP can enhance digital communication security. This project is a great tool for practice and experimentation with text classification and offers practical, hands-on insights into the ML model development process. 🔗 Try SpamShield here: https://omar-nahdi-sms-classifier.streamlit.app/

SpamShield: An ML powered Spam Detection Mode

🚢 Titanic Survivors Prediction with Machine Learning 🚀 I am thrilled to share my latest project where I successfully developed a machine learning model to predict the likelihood of survival for passengers aboard the Titanic. This project allowed me to delve into the fascinating intersection of data science and historical events. Project Highlights: 🔍 Data Exploration and Analysis: Conducted in-depth exploratory data analysis (EDA) to gain insights into the Titanic dataset. Explored key features, identified patterns, and visualized data distributions. 🧠 Feature Engineering: Implemented feature engineering techniques to enhance the predictive power of the model. Engineered relevant features, handled missing data, and transformed variables for optimal model performance. 🛠️ Model Development: Employed various machine learning algorithms, including decision trees, random forests, and logistic regression, to create and fine-tune the predictive model. Utilized cross-validation and grid search to optimize hyperparameters. 📊 Model Evaluation: Rigorously evaluated the model's performance using metrics such as accuracy, precision, recall, and F1 score. Employed confusion matrices and ROC curves to assess the model's robustness. 🔗 Implementation: Integrated the developed model into a user-friendly application, allowing for easy input of passenger information and providing predictions on survival probabilities. 📈 Results: Achieved impressive predictive accuracy and validated the model against test data. The project sheds light on the factors influencing survival rates during one of history's most iconic maritime disasters. 📚 Continuous Learning: Engaged in continuous learning throughout the project, staying abreast of the latest advancements in machine learning techniques and methodologies.