Rohan
Rohan
  • 9
  • 9 524
Presenter: A Multi-Agent AI Tool that creates beautiful presentations with voice-overs (tutorial)
Introducing Presenter: A Multi-Agent AI Tool that can:
-Create beautiful presentations for any given topic 🔥
-Render intuitive & visually appealing diagrams 🖼️ for slides when needed (using Mermaid)
-Write scripts for every slide 📜
-Render & View interactive presentations in HTML 💻 (using markdown-slides & reveal.js)
-Intuitive speaker view with scripts (reveal.js)
-Export presentations to PDF 🖨️ (using DeckTape)
-Generate audio narrations from scripts 🎙️ (using ElevenLabs)
-Render full video presentations 🎥 with all the slides and voiceover (using FFmpeg)
LlamaIndex Workflow is used for orchestrating the multi-agent setup
GitHub Repo: github.com/rsrohan99/presenter
Переглядів: 1 751

Відео

AI tour planner agent tutorial using LlamaIndex Workflow
Переглядів 77День тому
In this tutorial, we'll create a tour planner agent using LlamaIndex Workflow. Stack Used: - LlamaIndex workflow for orchestration. - SerpAPI for finding hotels, flights and places to visit GitHub Repo: github.com/rsrohan99/llamaindex-trip-planner
Streaming Events in LlamaIndex Workflows
Переглядів 1313 місяці тому
Let's see how we can stream intermediate events, e.g. token streaming or progress events to the client while using LlamaIndex Workflows. Code for the tutorial: github.com/rsrohan99/llamaindex-workflow-streaming-tutorial
Dynamic Few-shot prompting using LlamaIndex Workflows
Переглядів 5623 місяці тому
Dynamic Few-shot prompting is an intuitive alternative to finetuning LLMs where: - we can get consistent output for many use-cases - rapid iteration - takes effect right away - easy to implement In Dynamic Few-shot prompting, instead of putting a static list of examples in the prompt, we dynamically pull the best examples from our database based on the user query. GitHub Repo: github.com/rsroha...
Llama-Researcher: Build a research agent like GPT-Researcher using the new LlamaIndex workflows
Переглядів 8834 місяці тому
In this tutorial, we'll create LLama-Researcher using LlamaIndex workflows, inspired by GPT-Researcher. GitHub: github.com/rsrohan99/llama-researcher Stack Used: - LlamaIndex workflows for orchestration - Tavily API as the search engine api - Other LlamaIndex abstractions like VectorStoreIndex, PostProcessors etc.
Automated Metadata Extraction & Filtering Pipeline using LlamaExtract and LlamaIndex Auto Retriever
Переглядів 3,1 тис.5 місяців тому
LlamaExtract by LlamaIndex is a fully-managed service to extract data from unstructured and complex files. One usecase for LlamaExtract is metadata extraction. In this tutorial, we'll see how to use LlamaExtract to automate the entire process of metadata extraction, creating vector index and retrieve nodes via automatically generated filters from user query. We'll use Auto Retriever abstraction...
AI Diagram generator using LlamaIndex & Vercel with partial parsing to show diagrams being built
Переглядів 9989 місяців тому
Previous tutorial on streaming intermediate events: ua-cam.com/video/JOM8WgmNCvI/v-deo.htmlsi=S66jkBZtByM2LqcK In this tutorial, we'll build a cool AI diagram generator using LlamaIndex and Vercel AI SDK. GitHub Repo: github.com/rsrohan99/ai-diagram-generator It takes the topic as input and creates a comprehensive and detailed diagram of that topic. It uses LlamaIndex pydantic program mode for ...
Use LlamaIndex and Vercel AI SDK to stream intermediate events to the frontend for better UX
Переглядів 7159 місяців тому
Sending intermediate events in RAG is crucial for best user experience 🚀 Let's see how to use LlamaIndex Instrumentation module, with server-sent events and Vercel AI SDK's StreamData feature to properly stream intermediate events to the frontend. Full tutorial under 3 minutes 🔥 Repo: github.com/rsrohan99/rag-stream-intermediate-events-tutorial
Fully Local Chat-With-PDF App tutorial under 2.5 minutes 🚀 Using LlamaIndex TS, Ollama, Next.JS
Переглядів 1,4 тис.9 місяців тому
Fully local, open-source chat-with-pdf app tutorial under 2.5 minutes. It supports chat with pdf fully locally using Ollama to run both embed and language models. It also has preview feature that auto scrolls the PDF to the page that contains the top retrieved hunk from the retriever. Stack used: LlamaIndex TS for RAG Ollama NextJS with server action Phi2 and Nomic AI emedding models using Olla...

КОМЕНТАРІ

  • @imyadavcodes1418
    @imyadavcodes1418 5 днів тому

    Nice one!, what tool did you use for creating the architecture diagram?

  • @sagartamang0000
    @sagartamang0000 15 днів тому

    Hey, that was really helpful, thank you so much!

  • @BrettKromkamp
    @BrettKromkamp 3 місяці тому

    Good overview and thanks for providing the accompanying GitHub repository. Appreciated. Subscribed to your channel.

  • @Htyagi1998
    @Htyagi1998 4 місяці тому

    Do you have any udemy or some other course where you are explaining all these stuff in detail?

  • @anumulavirajith9927
    @anumulavirajith9927 5 місяців тому

    Within the MetaDataInfo class, if I have to mention about a metadata field that takes categorical values, how do I do that? And how do I ensure the model understands to take only those values and nothing else, Eg: status is a metadata which takes (1 -important, 2-not important, 3-not relevant.) and document objects have the status metadata taking these values so it should be able to pick only those 3 options. Thanks

  • @rsrohan99
    @rsrohan99 5 місяців тому

    Demo notebook: github.com/rsrohan99/tutorial-notebooks/blob/main/llama-index/llama-extract-metadata-extraction-tutorial.ipynb

  • @pranavsingh2592
    @pranavsingh2592 5 місяців тому

    github repo link?

  • @BowenChen-sh3sz
    @BowenChen-sh3sz 8 місяців тому

    Thank you so much for the tutorial, I'm seeing create-llama has a significant update recently. Any chances you could update this code to reflect the latest changes?

    • @rsrohan99
      @rsrohan99 8 місяців тому

      Thanks. I'll definitely try to update it

    • @BowenChen-sh3sz
      @BowenChen-sh3sz 8 місяців тому

      @@rsrohan99 Thank you so much !

  • @pascalgula
    @pascalgula 9 місяців тому

    amazing!

  • @causalinference4176
    @causalinference4176 9 місяців тому

    good job, thank you

  • @ousheshharadhun3773
    @ousheshharadhun3773 9 місяців тому

    whats the backend written in inherently? Fastapi or django or flask

  • @markr9640
    @markr9640 9 місяців тому

    I like brevity... but dude, you went so fast, I nearly fell over. I would have liked to see you use the application, once built too. Anyway, keep up the good work, you will soon have many subs 👍

    • @rsrohan99
      @rsrohan99 9 місяців тому

      Thanks mate, appreciate the feedback. Will keep that in mind 👍