LangChain In Action: Voice of Customer Modeling With Zapier

Поделиться
HTML-код
  • Опубликовано: 3 дек 2024

Комментарии • 16

  • @zagoguic
    @zagoguic Год назад

    You are bringing such great content for RUclips about LangChain and GPT. Keep going! Hope you are enjoying it!

    • @rabbitmetrics
      @rabbitmetrics  Год назад

      Appreciate the comment! 🙏 Thanks for watching

  • @jayhu6075
    @jayhu6075 Год назад

    What a great explanation. As a beginner in AI I learn a lot from this channel. Many thanks.

  • @adiyogi-thefirstguru5144
    @adiyogi-thefirstguru5144 Год назад

    Thanks for the video.
    Few questions
    1. How can I train with Database and build chatbot.
    2. Is it possible to give input the whole website and will it get trained?
    3. Is it possible to train the financial data like spending, goals, budgeting etc.. so that it'll automatically suggest the financial goal plannings?
    4. Is it possible to train for stock market trading purpose?

  • @LennartBiesel
    @LennartBiesel Год назад

    This is the coolest stuff I just randomly walked across. Everything about the idea and the real world application is just amazing. This is the only acutally usabale and scalable AI business video stuff on all of youtube the rest is just clickbaite

    • @rabbitmetrics
      @rabbitmetrics  Год назад

      Thank you! I appreciate the comment and I’m glad you found the content usable

  • @unclecode
    @unclecode Год назад +1

    Thanks for the content. While this video showcases Langchain agents, I'd like to focus on the case study it presents - sending personalized emails to people who have left us reviews. I question the need for mixing AI agents with Zapier NLA for deterministic tasks like emailing 5-star reviewers. By engineering well-structured prompts, we could use GPT to generate the email content and directly utilize the Mailgun API, potentially saving both computational and financial resources. As powerful LLMs like GPT-4 continue to evolve, it appears that reliance on services like Zapier NLA, which converts human language queries into structured information, is diminishing. In this instance, I struggle to see the benefits of mixing GPT-4 with the Zapier NLA API. I do understand that the main objective of this video is to demonstrate the capabilities of Langchain Agents and I have no objections to that. I'd appreciate hearing your thoughts on these points.

    • @rabbitmetrics
      @rabbitmetrics  Год назад +1

      Thanks for watching! Yes, you can cut out Zapier and built a custom personalization agent to interact with any API. Let's say I wanted to send out personalized emails with Klaviyo. Then I would interface directly with the Klaviyo API as this would allow me to utilize the full functionality of the API. I also expect a custom agent would be more stable. I will dive into this in future videos.

  • @fabianaltendorfer11
    @fabianaltendorfer11 Год назад

    Awesome, thanks!

  • @RedShipsofSpainAgain
    @RedShipsofSpainAgain Год назад

    How much did it cost (OpenAI credits) to build this?

  • @khaledarja9239
    @khaledarja9239 Год назад

    great content man

  • @jalalabadass
    @jalalabadass Год назад

    How did it retrieve the email addresses of the customers who reviewed the product?

    • @rabbitmetrics
      @rabbitmetrics  Год назад

      Names and email addresses were simulated with Faker in the video. Normally you'd be able to get this info from the CRM