Ravi Kumar Blog Posts

Machine learning is a subset of artificial intelligence that uses algorithms (sets of instructions given to the computer) trained on datasets to create models capable of performing tasks such as categorizing images, analyzing data, or predicting price fluctuations. ML Lifecycle: Types of ML: Supervised learning: These are trained on the labeled dataset. They learn to map input features to targets based on labeled training data. There are two main types of supervised learning: Regression algorithm learns to predict continuous values based on input features. Classification algorithm learns to assign input data to a specific category or class based on input features. The output labels in classification are discrete values. Unsupervised learning: It learns to recognize patterns in data without being explicitly trained using labeled examples. The goal is to discover the underlying structure or distribution in the data. Clustering algorithms group similar data points together ba...

 Retrieval-Augmented Generation (RAG): In simple terms, RAG is a technique that uses AI to improve the accuracy of text generation. RAG is a framework that combines large language models like GPT-3 with external data sources. It works by retrieving relevant information from databases or docs and then using it to generate better responses. RAG applications use an LLM to generate responses and retrieve information from a knowledge base to improve the accuracy of the answers. Most companies building RAG applications for customer support, education, content recommendations, and entertainment etc. For example, imagine you asked a RAG-powered chatbot, “ What’s the capital of India? ” The chatbot would first quickly search an online database to find that “ The capital of India is New Delhi. ” Then, using that retrieved information, the chatbot would respond, “ The capital of India is New Delhi. ” So, in a nutshell, a RAG system:  Retrieve the relevant data Use that data to augment ...

 I have posted a simple usecase on AI Agents earlier and was talking about the extraction of invoices data from Gmail and store it in the Google Sheets. Here is the link to recap of this implementation: Click Now this is another usecase with a bit of complexity added. From this use case, we are going to search and get the LinkedIn profile URL for the asked profile captured in the google sheet. If there are more than one URLs then we have a manual intervention to finalize the right URL. In this manual intervention process, we send the list of all URLs in an email to take action from the gmail. This agent runs manually based on the list of profiles added to the sheet and is not automatic trigger like in the previous usecase. The whole picture of this implementation using Relay.app Here are the steps to be followed for this usevcase implementation by using Relay.app  1) Login into Relay.app site with the user credentials 2) Create a trigger with the below setup Here is the sheet ...

An AI, or Artificial Intelligence, is essentially a computer program designed to perform tasks that typically require human intelligence.  Think of AI as a very sophisticated tool that can process massive amounts of information and find insights or solutions much more quickly than a human brain. It's like having an incredibly smart assistant that can quickly analyze data, recognize complex patterns, and provide intelligent responses or actions.  AI is more than just algorithms. It’s a process that needs careful planning, the right data, and a skilled team to make things work. Examples of AI: a) Optical character recognition (OCR): Uses AI to extract text and data from images and documents  b) Voice assistants: Like Siri and Alexa, which use AI technology  c) Customer service chatbots: Help users navigate websites  This is what an end-to-end AI project looks like: Define the Problem:   Before jumping into AI, it’s important to clearly understand what proble...

Hugging Face is a platform that provides tools and libraries for natural language processing (NLP), machine learning, and artificial intelligence. It is known for its Transformers   library, which offers pre-trained models for tasks such as text classification, translation, question answering, and more. Hugging Face also hosts a model hub where users can share and discover pre-trained models. Hugging Face is useful in various domains, including: Natural language processing (NLP): Provides tools and pre-trained models for tasks like text classification, translation, sentiment analysis, and question answering. Machine learning research: Offers state-of-the-art models and datasets for researchers to experiment and innovate. Data science and analytics: Helps data scientists analyze and interpret large volumes of text data. AI application development: Enables developers to build AI-powered applications with ease using pre-trained models and APIs. A very simple exercise using the google...