Still searching with keywords?
Frustrated with the results?
Struggling to search across all your data?
SerenityGPT indexes all the content you want, lets you ask natural language questions—just like asking a teammate— and get accurate, unified, real-time responses.
We’ll even set it up for you!
Book your demo today
Already with us
How it works
I am getting a blank screen when I log in to the web dashboard.
Book a Demo
John Smith
New team member
Serenity
Response
Please try the following:
— clear cookies by clicking the lock in the address bar and log in again
— open the page in an incognito browser window and log in again
— clear cookies by clicking the lock in the address bar and log in again
— open the page in an incognito browser window and log in again
Common FAQ
Blank screen issue. #support 2023-07-08
What We Offer
Auto-pilot for support
Respond to customer queries with correct answers from your docs/KB/tickets.
Co-pilot for support
Surface the right info and phrase the answer for support reps to respond.
Enterprise search
Find the right information across your company's data as if you were talking to a colleague who knows it all.
Smart Search
Let your customers immediately get answers to “how to" questions
1
2
3
4
How we are different
Serenity is fully installed and configured by us unlike a self-service toolkit or an API. We can manage it for you or turn over to your IT.
A turnkey solution
Your data remains on your servers. Also, you are not tied to any cloud provider.
On-prem
Would you like the relevant information to automatically show up in Freshdesk, or Slack, or Teams? We configure integrations for you.
Integrated with your workflows
Benefits
Productivity and cost savings
— cut support costs by 50%
— provide automated support where you had none
— make your teams more efficient
— provide automated support where you had none
— make your teams more efficient
Self-learning search
All your data turned into
an up-to-date knowledge
base without creating one
an up-to-date knowledge
base without creating one
A secure enterprise-ready solution
Your data does not leave your servers
Happier teams
Empowered to find the information they need
Security of customer data
Our standard approach is to keep all data on customer servers.
Historical and real-time data loading on your servers
Local processing
Users only search over the data they have access to
Permissions
GPT-4 via Azure to phrase the answer given the relevant documents with removed PII
Compliant
Optional
FAQ
ChatGPT is awesome but on its own it does not help much with enterprise search. First, ChatGPT does not know anything about your data. Second, even if it did (think fine-tuning), it would be prone to providing incorrect answers (hallucinations) and often unable to link its answers back to the sources of information. The approach that overcomes these difficulties is called Retrieval Augmented Generation (RAG). RAG consists of two components: a "retriever" that finds information relevant to the question and a large language model (e.g., ChatGPT) that answers the question based on the information provided by the retriever. In addition to that, enterprise search has other dimensions that need to be implemented in a RAG system:
— permissions: users search only over the content they are permissioned to see
— "faceted" search where you can filter / prioritize by categories configured based on your data such as the data source (e.g., knowledge base articles may take precedence over tickets), the freshness of the document (when was it last modified), the author of the document, etc
— ability to learn from feedback both implicit and explicit (e.g., upvote/downvote documents and specify the right document if it's not found)
More details about the components of Serenity are in our whitepaper above.
— permissions: users search only over the content they are permissioned to see
— "faceted" search where you can filter / prioritize by categories configured based on your data such as the data source (e.g., knowledge base articles may take precedence over tickets), the freshness of the document (when was it last modified), the author of the document, etc
— ability to learn from feedback both implicit and explicit (e.g., upvote/downvote documents and specify the right document if it's not found)
More details about the components of Serenity are in our whitepaper above.
LangChain provides a Question Answering pipeline: https://python.langchain.com/docs/use_cases/question_answering/ By default it uses OpenAI's embeddings and ChatGPT and sends all of the data to OpenAI for "indexing". Every data source need to be loaded and they provide a number of loaders. The issues relevant to enterprise search (see"Why Can't ChatGPT do this" above): permissions, faceted search, learning from feedback are not part of the pipeline. In short, LangChain may help but it won't be easy.
LangChain uses ChatGPT by default. While it is possible to use open source models, care needs to be taken to select the best models for your use case. Any system not adapted to your specific data will not work well with terms specific to your company, such as industry-specific jargon, project/product names, abbreviations, etc.
LangChain uses ChatGPT by default. While it is possible to use open source models, care needs to be taken to select the best models for your use case. Any system not adapted to your specific data will not work well with terms specific to your company, such as industry-specific jargon, project/product names, abbreviations, etc.
Fine-tuning will not help with hallucinations and the LLM may not be able to link its answers back to the sources it used to generate them (RAG is an approach that does not suffer from these problems). There are other issues with fine-tuning an LLM: it works best when there is a LOT of data, it may get expensive, and it is difficult to update (e.g., add one new document or a new version of a document).
Like with any other technology investment "buy vs build" is an important decision each organisation needs to make for themselves. We have many years of experience building search and AI engines for different use cases, and we continually improve our technology based on the work we do with multiple clients. Having enough in-house expertise to build, maintain, and update the solution is usually much more expensive than buying.
hello@serenitygpt.com
SerenityGPT Limited
London, United Kingdom
Whitepaper
Whitepaper
© 2024, SerenityGPT
Cookies managing
Cookie Settings
Cookies necessary for the correct operation of the site are always enabled.
Other cookies are configurable.
Other cookies are configurable.