Avratanu Biswas’ Post

text to video

The Grounding with Google Search family continues to expand! 🐣 Take your Gemini model outputs to the next level and ground with Google Search to ensure freshness & accuracy from the web 🔍 Try it now in Google AI Studio! 🤩 #rag #grounding #googlecloud #groundingwithgooglesearch #geminiera

Hello everyone, have you already tried Google AI Studio https://lnkd.in/eca44dyT. What are you waiting for? It has spectacular features that will surprise you. In the future I will bring some ideas using these tools to introduce the use of AI in automated testing. #WebDevelopment #JavaScript #FrontendDevelopment #Software #Coding #TechTips #GoogleCloud

Asking Mistral AI's LLMs to perform actions (or at least to have actions performed) is now possible with the superb Function Calling API. 👩💻 Discover this API in the following article and get hands-on with the code with a ready-to-use Flask app! 👉 Article: https://lnkd.in/dNHqiJDy 💡In the demo, there's a home automation application controlled via the large Mistral model. Feel free to share the article with developers around you. 📚 cc: Sophia Yang, Ph.D. #ai #llm #lam #mistral

Easier tools to build AI web apps than Streamlit

A RAG recipe for a contextualized chatbot app - "text-embedding-3-small" OpenAI model for creating vector embeddings, "Pinecone" vector database indexes for storing the embeddings and OpenAI's "chatbot-3.5-turbo" for getting the answers based on the augmented contextualized query. This kind of applications have lots of use-cases for enterprises. I used Gradio to create a quick demo of the RAG application which you may view at the following link: https://lnkd.in/gVZTsGnm #rag #semanticsearch #llms #deeplearning #statsomlvignette #statsomlai #machinelearning

Build AI Apps in a Flash with Gradio's Reload Mode 🚀 This post explores Gradio's reload mode for rapid AI app development: Faster Iteration: Instantly see UI changes without server restarts. ⚡️ Selective Reloading: Avoids unnecessary model reloading for efficiency. 🔍 Example: Document Analyzer App 📄 Users upload documents & ask questions. 🙋♀️ Answers come in natural language. 🗣️ Powered by Hugging Face Inference API (free & no GPU required!). 💻 Get Started Today! 🚀 This short post provides a concise overview of Gradio's reload mode benefits and a practical example of building a document analyzer app. #AIapps #GradioReload #DocumentAnalyzer #HuggingFace #NoGPU #ai https://lnkd.in/gziyy2RV

🤯 My mind is blown. I built & deployed an in-browser note-taking app in 3 HOURS using Bolt (https://bolt.new)! What exactly is Bolt? Bolt is an AI tool that can build and deploy apps from scratch. 🔨 Here's some of the features I asked Bolt to build:  - Browser-based note-taking with AI writing assistance (use your claude api key) - Saves notes locally for privacy - Support general notes & twitter thread creation - Minimal UI with folders & simple shortcuts If you want to play around with this, here's the link: https://lnkd.in/gCH4DYAD

Congratulations to our very own Sandesh Chapagain for winning the Kearney GEN AI Innovation Storm ⚡: finding the best GenAI use case to make consulting faster, better, cheaper, and more fun. With Hunter Dunbar's guidance, Sandesh created 'Kearney Prompt Helper', a GenAI tool that is designed to streamline the creation, storage and execution of of grounded prompt templates, facilitating development of a rich prompt repository that can be used in diverse consulting domains. Sandesh also delivered a full function web app! Claps all around! 👏 #winwithdata

AI/ML Integration: Streamlined Integration: Libraries like TensorFlow.js and PyTorch.js are making it easier to integrate machine learning models directly into web apps, leading to more interactive and intelligent user experiences.