Faiqa Rafiq@Faiqa

The weather AI app is a multifaceted platform designed to offer more than just standard weather updates. It integrates real-time weather data for any location, fetching details like temperature, humidity, wind speed, and weather descriptions using the OpenWeatherMap API. Additionally, the app retrieves an image of the location from the Unsplash API, providing a visual representation of the place for a richer user experience. One of the unique aspects of the app is its ability to generate personalized suggestions based on the current weather. For instance, depending on the temperature and weather conditions (rain, snow, clear skies), the app suggests suitable clothing, activities, and even whether it's a good day for a picnic. These recommendations are generated using Hugging Face’s GPT-2 model, adding a layer of AI-driven decision-making to enhance user interaction. To further elevate the experience, the app converts these recommendations into audio using Hugging Face's FastSpeech2 model. This feature allows users to listen to the recommendations instead of reading them, making the app accessible and interactive for various users. Overall, the app combines real-time weather data, city imagery, smart suggestions, and voice-enabled features to deliver an engaging and informative weather solution.

An IBM website chatbot is an advanced AI-powered virtual assistant that plays a pivotal role in enhancing the user experience on IBM’s digital platforms. Designed with cutting-edge technology, the chatbot is integrated into IBM’s website to provide real-time support, guidance, and information to users, making their interaction with the site more efficient and user-friendly. This chatbot is not just a simple query-response system; it is a comprehensive tool that leverages sophisticated AI models, such as the IBM Granite 13B, to deliver a wide range of functionalities tailored to meet the needs of various users. ### Purpose and Functionality The primary purpose of the IBM website chatbot is to assist users in navigating the website, finding relevant information, and resolving issues without the need for human intervention. This is particularly important for large, complex websites like IBM’s, where users might find it challenging to locate specific information or troubleshoot issues on their own. The chatbot serves as a virtual guide, leading users to the right resources, answering frequently asked questions (FAQs), and providing step-by-step assistance for more complicated queries. The chatbot's functionality extends beyond just answering questions. It is designed to handle a variety of tasks that enhance the user experience, such as: - *Website Navigation Assistance*: The chatbot helps users find the information they need by guiding them through different sections of the website. Whether a user is looking for specific product details, support documentation, or company information, the chatbot can quickly direct them to the right page. - *Customer Support*: For users seeking support with IBM products or services, the chatbot offers troubleshooting tips and solutions. It can address common issues directly or escalate more complex problems to human support agents when necessary.

SuperHuman AI: Automating web-based tasks with AI-driven intelligence for improved efficiency Objective: SuperHuman AI is designed to automate complex activities over web browsers based on user-defined objectives. By leveraging advanced AI models for vision and text intelligence, the project focuses on streamlining tasks such as job applications, form-filling, and data entry. It reduces manual effort and human error, enabling faster execution and better user productivity. Key Features: Visual Intelligence: Powered by Llama 3.2 11B Multimodal Vision Instruct Model, capable of processing visual data to automate complex tasks that require an understanding of on-screen elements. Automation through Selenium: Uses Python Selenium for navigating and interacting with web pages, understanding DOM elements, and mimicking user interactions. Advanced RAG Pipeline: Integrates Vectara and Chroma DB for advanced retrieval-augmented generation (RAG) to provide real-time, accurate answers and suggestions during automation. AI Agent with CrewAI: Built using CrewAI for orchestrating complex tasks in sequence, with the ability to break down objectives into manageable steps and execute them efficiently. AI Agent Steps: Initiates the browser and navigates to the target website. Utilizes Selenium to analyze and understand the site's DOM elements. Visual intelligence processes visual data for more advanced activities (e.g., interpreting dynamic content, etc.). Gathers the user’s overall objective and breaks it into implementable steps. Executes each step while adapting to the dynamic web environment. Provides analytics and feedback to the user at the end of the task. Core Technologies: AI Agent: CrewAI LangGraph AI Tools: Llama 3.2 11B Multimodal Vision Instruct Model Llama 3.2 3B Lightweight Text Model Advanced RAG Pipeline Use-cases: LinkedIn Job Automation Google Forms Filler KYC Processing Data Entry Automation