docs-pages icon indicating copy to clipboard operation
docs-pages copied to clipboard
verified publisher icon holochain

Implementation of letmeexplain.ai

Open letmeexplainAI opened this issue 10 months ago • 7 comments

This PR integrates the LetMeExplain.ai chatbot into the Holochain developer docs to provide AI-powered responses to user queries. The chatbot is loaded via an external script and connects to a Large Language Model (LLM) to generate responses. The LLM is currently trained only on the Holochain Whitepaper.

Technical Details:

  • The script (lme_chatbot_widget.js) is hosted on https://dashboard.letmeexplain.ai.
  • It initializes a chatbot widget with a unique orgId to identify the client.
  • The chatbot interacts with an LLM to respond to user queries.

Security and Privacy:

  • The script does not collect or store sensitive user data.

Support:

For access to the LetMeExplain.ai client dashboard or technical support, please contact:

letmeexplainAI avatar Feb 04 '25 07:02 letmeexplainAI

Hey @letmeexplainAI , interesting proposal, and it's something we've contemplated adding to our dev docs already.

Right now our first priority is getting actual documentation written, then training an LLM on it. The white paper already reflects a slightly out-of-date model of the living code, and will get more stale as time goes on. Can you tell me more about your product? Big questions are:

  1. Will it progressively update its model as our documentation improves? (I assume it's doing RAG or fine-tuning or building up a huge context window or something)
  2. What base model does it use?
  3. What are your pricing plans for open-source projects stewarded by non-profits like ours?

pdaoust avatar Feb 05 '25 19:02 pdaoust

Hi,

Thank you for reaching out. Below are the answers to your questions regarding our product.

Progressive Model Updates: Yes, LME is built on a RAG (Retrieval-Augmented Generation) architecture using open-source technologies on AWS. Documentation updates can be managed through the client dashboard, where you can manually add new content. Additionally, we periodically scan existing URLs to ensure your knowledge base remains up to date. 2.

Base Model: We use OpenAI's 4o-mini as the foundational model for LME. 3.

*Pricing for Non-Profits: *We are an AI Lab and love nonprofits.

  • The whitepaper-based chatbot and general Q&A features are available for free.
    • If your documentation dataset expands significantly or user traffic increases, you have the option to continue using the free bot on the whitepaper or transition to an affordable paid plan that primarily covers our operational costs.

Please let us know if you have any further questions or if you'd like a demo. Thanks, Owais

On Wed, Feb 5, 2025 at 7:18 PM pdaoust @.***> wrote:

Hey @letmeexplainAI https://github.com/letmeexplainAI , interesting proposal, and it's something we've contemplated adding to our dev docs already.

Right now our first priority is getting actual documentation written, then training an LLM on it. The white paper already reflects a slightly out-of-date model of the living code, and will get more stale as time goes on. Can you tell me more about your product? Big questions are:

  1. Will it progressively update its model as our documentation improves? (I assume it's doing RAG or fine-tuning or building up a huge context window or something)
  2. What base model does it use?
  3. What are your pricing plans for open-source projects stewarded by non-profits like ours?

— Reply to this email directly, view it on GitHub https://github.com/holochain/docs-pages/pull/521#issuecomment-2637824419, or unsubscribe https://github.com/notifications/unsubscribe-auth/BPAVMTZCGYKWTSHMZUNKJ332OJPZNAVCNFSM6AAAAABWN3L4T2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMZXHAZDINBRHE . You are receiving this because you were mentioned.Message ID: @.***>

letmeexplainAI avatar Feb 07 '25 09:02 letmeexplainAI

Hi there,

Following up on the previous email, I hope the responses provided addressed your questions adequately.

Please let me know if you have any further questions or require additional information. I'd be happy to schedule a call/demo at your convenience.

Best Regards, Owais

On Fri, Feb 7, 2025 at 9:37 AM LetMeExplain AI @.***> wrote:

Hi,

Thank you for reaching out. Below are the answers to your questions regarding our product.

Progressive Model Updates: Yes, LME is built on a RAG (Retrieval-Augmented Generation) architecture using open-source technologies on AWS. Documentation updates can be managed through the client dashboard, where you can manually add new content. Additionally, we periodically scan existing URLs to ensure your knowledge base remains up to date. 2.

Base Model: We use OpenAI's 4o-mini as the foundational model for LME. 3.

*Pricing for Non-Profits: *We are an AI Lab and love nonprofits.

  • The whitepaper-based chatbot and general Q&A features are available for free.
    • If your documentation dataset expands significantly or user traffic increases, you have the option to continue using the free bot on the whitepaper or transition to an affordable paid plan that primarily covers our operational costs.

Please let us know if you have any further questions or if you'd like a demo. Thanks, Owais

On Wed, Feb 5, 2025 at 7:18 PM pdaoust @.***> wrote:

Hey @letmeexplainAI https://github.com/letmeexplainAI , interesting proposal, and it's something we've contemplated adding to our dev docs already.

Right now our first priority is getting actual documentation written, then training an LLM on it. The white paper already reflects a slightly out-of-date model of the living code, and will get more stale as time goes on. Can you tell me more about your product? Big questions are:

  1. Will it progressively update its model as our documentation improves? (I assume it's doing RAG or fine-tuning or building up a huge context window or something)
  2. What base model does it use?
  3. What are your pricing plans for open-source projects stewarded by non-profits like ours?

— Reply to this email directly, view it on GitHub https://github.com/holochain/docs-pages/pull/521#issuecomment-2637824419, or unsubscribe https://github.com/notifications/unsubscribe-auth/BPAVMTZCGYKWTSHMZUNKJ332OJPZNAVCNFSM6AAAAABWN3L4T2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMZXHAZDINBRHE . You are receiving this because you were mentioned.Message ID: @.***>

letmeexplainAI avatar Feb 10 '25 13:02 letmeexplainAI

Thanks! Every question answered except one: what are your pricing plans? I don't see that on your website.

I'd also like to know: how long have you been in business?

Thanks!

pdaoust avatar Feb 10 '25 19:02 pdaoust

Hi,

Sorry I thought I replied to this. Apologies for the delay.

Here's a link to our pricing page.

We have been in business for 2 years now.

Regards Owais

On Mon, Feb 10, 2025 at 7:41 PM pdaoust @.***> wrote:

Thanks! Every question answered except one: what are your pricing plans? I don't see that on your website.

I'd also like to know: how long have you been in business?

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/holochain/docs-pages/pull/521#issuecomment-2649062651, or unsubscribe https://github.com/notifications/unsubscribe-auth/BPAVMT3PAFZLBZSPLX7DNHT2PD6FBAVCNFSM6AAAAABWN3L4T2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBZGA3DENRVGE . You are receiving this because you were mentioned.Message ID: @.***>

letmeexplainAI avatar Feb 22 '25 12:02 letmeexplainAI

Here's the link to our pricing page: https://www.letmeexplain.ai/pricing/

On Sat, Feb 22, 2025 at 12:31 PM LetMeExplain AI @.***> wrote:

Hi,

Sorry I thought I replied to this. Apologies for the delay.

Here's a link to our pricing page.

We have been in business for 2 years now.

Regards Owais

On Mon, Feb 10, 2025 at 7:41 PM pdaoust @.***> wrote:

Thanks! Every question answered except one: what are your pricing plans? I don't see that on your website.

I'd also like to know: how long have you been in business?

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/holochain/docs-pages/pull/521#issuecomment-2649062651, or unsubscribe https://github.com/notifications/unsubscribe-auth/BPAVMT3PAFZLBZSPLX7DNHT2PD6FBAVCNFSM6AAAAABWN3L4T2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBZGA3DENRVGE . You are receiving this because you were mentioned.Message ID: @.***>

letmeexplainAI avatar Feb 22 '25 12:02 letmeexplainAI

Hi,

Just checking if you've received my response.

Regards Owais

On Sat, Feb 22, 2025 at 12:32 PM LetMeExplain AI @.***> wrote:

Here's the link to our pricing page: https://www.letmeexplain.ai/pricing/

On Sat, Feb 22, 2025 at 12:31 PM LetMeExplain AI @.***> wrote:

Hi,

Sorry I thought I replied to this. Apologies for the delay.

Here's a link to our pricing page.

We have been in business for 2 years now.

Regards Owais

On Mon, Feb 10, 2025 at 7:41 PM pdaoust @.***> wrote:

Thanks! Every question answered except one: what are your pricing plans? I don't see that on your website.

I'd also like to know: how long have you been in business?

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/holochain/docs-pages/pull/521#issuecomment-2649062651, or unsubscribe https://github.com/notifications/unsubscribe-auth/BPAVMT3PAFZLBZSPLX7DNHT2PD6FBAVCNFSM6AAAAABWN3L4T2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBZGA3DENRVGE . You are receiving this because you were mentioned.Message ID: @.***>

letmeexplainAI avatar Feb 28 '25 08:02 letmeexplainAI

Still willing to keep letmeexplain.ai in mind, but I'm closing cuz we're not ready to choose an AI service and integrate yet.

pdaoust avatar Sep 19 '25 16:09 pdaoust