cancel
Showing results for 
Search instead for 
Did you mean: 
Generative AI
Explore discussions on generative artificial intelligence techniques and applications within the Databricks Community. Share ideas, challenges, and breakthroughs in this cutting-edge field.
cancel
Showing results for 
Search instead for 
Did you mean: 

Forum Posts

Sujitha
by Databricks Employee
  • 38465 Views
  • 22 replies
  • 48 kudos

Databricks Announces the Industry’s First Generative AI Engineer Learning Pathway and Certification

Today, we are announcing the industry's first Generative AI Engineer learning pathway and certification to help ensure that data and AI practitioners have the resources to be successful with generative AI. At Databricks, we recognize that generative ...

Screenshot 2024-01-24 at 11.32.01 PM.png
  • 38465 Views
  • 22 replies
  • 48 kudos
Latest Reply
Darshan137
New Contributor II
  • 48 kudos

Dear Certifications TeamI have completed full Generative AI Engineering Pathway, so I received module wise knowledge badge but I didn't received the overall certificate which mentioned in description which is Generative AI Engineer with one Star. Req...

  • 48 kudos
21 More Replies
gendg
by New Contributor
  • 6219 Views
  • 1 replies
  • 0 kudos

Mlflow.evaluation fails to generate score

The execution of code stucks when evaluation of data start.  eval_df = pd.DataFrame(    {        "inputs": [            "What is MLflow?",            "What is Spark?",        ],        "ground_truth": [            "MLflow is an open-source platform f...

Generative AI
Databricks
GenAI
ML
mlflow
  • 6219 Views
  • 1 replies
  • 0 kudos
Latest Reply
mark_ott
Databricks Employee
  • 0 kudos

The issue described—a Databricks notebook getting "stuck" during the evaluation phase using mlflow.evaluate—is most likely related to environment setup, model compatibility, or limitations with the mlflow.pyfunc.log_model and the evaluation utilities...

  • 0 kudos
SandipCoder
by New Contributor II
  • 5906 Views
  • 1 replies
  • 0 kudos

I need a sample code or process which will help us to dynamically select the prompt template

We need a sample code or process which will help us to dynamically select the prompt template based on the prompt given as an input through the model legacy serving endpoint

Generative AI
GenAI
rag
  • 5906 Views
  • 1 replies
  • 0 kudos
Latest Reply
mark_ott
Databricks Employee
  • 0 kudos

To dynamically select a prompt template in Databricks based on the input prompt received through a legacy model serving endpoint, you can implement a Python function that maps incoming prompts to specific templates. This often involves using conditio...

  • 0 kudos
narenderkumar57
by Databricks Partner
  • 8920 Views
  • 1 replies
  • 0 kudos

How to perform combined search on structured and unstructured data in databrick using RAG or other

I created a RAG application in databricks which performs the following steps:1. Extract text from PDF files2. Prepare embeddings on extracted text and create vector search index3. Create a LLM model and served the model which can answer question base...

  • 8920 Views
  • 1 replies
  • 0 kudos
Latest Reply
mark_ott
Databricks Employee
  • 0 kudos

You can achieve combined retrieval across both PDF-extracted unstructured data and multiple columns from structured Delta tables in Databricks, but there are important considerations and available patterns to optimize this workflow for your RAG appli...

  • 0 kudos
SandipCoder
by New Contributor II
  • 5028 Views
  • 1 replies
  • 0 kudos

Issue with Multi agent supervisor based agentic framework

I  have three agents in my multi agent framework (code attached) Supervisor : This is the main controller    "Genie":           #this is a genai agent    "Coder":          #this is a re-act agent created on the fly    "Weather":      #this is an exis...

  • 5028 Views
  • 1 replies
  • 0 kudos
Latest Reply
mark_ott
Databricks Employee
  • 0 kudos

Your multi-agent framework has several issues affecting agent iteration control and error handling. Here’s a lined explanation and practical suggestions for each problem: Agents not stopping at one iteration Agents consistently reaching the max iter...

  • 0 kudos
MattB
by New Contributor III
  • 4991 Views
  • 1 replies
  • 0 kudos

AI/BI Genie - Components

I am almost certain that AI/BI Genie is using Azure OpenAi under the hood. Does anyone knows if Langchain has been used too ?

  • 4991 Views
  • 1 replies
  • 0 kudos
Latest Reply
mark_ott
Databricks Employee
  • 0 kudos

AI/BI Genie is confirmed to use Azure OpenAI under the hood, as it leverages generative AI models for translating natural language into analytical queries and producing business intelligence insights. Additionally, there is strong evidence that Langc...

  • 0 kudos
Ulfzerk
by New Contributor
  • 5140 Views
  • 1 replies
  • 1 kudos

Custom sentence transformer for indexing

Hi! i would like to use my own sentence transformer to create a vector index. It is not a problem using mlflow sentence-transformer flavour, it works fine with: mlflow.sentence_transformers.log_model( model, artifact_path="model", signatu...

  • 5140 Views
  • 1 replies
  • 1 kudos
Latest Reply
mark_ott
Databricks Employee
  • 1 kudos

To use a custom MLflow pyfunc model for sentence-transformers with preprocessing, you need to comply with the expected interface of mlflow.pyfunc.PythonModel, especially the predict method. The method signature, data handling, and serialization are k...

  • 1 kudos
smferro54epam
by Databricks Partner
  • 4321 Views
  • 1 replies
  • 0 kudos

Resolved! None of the Connect to the custom MCP server examples work

hi! I'm following through this documentation >> https://docs.databricks.com/aws/en/generative-ai/mcp/custom-mcp?language=Agent+code+%28service+principal%29 I have successfully deployed the app in Databricks and can use it from the playground and from...

  • 4321 Views
  • 1 replies
  • 0 kudos
Latest Reply
Louis_Frolio
Databricks Employee
  • 0 kudos

Greetings @smferro54epam , the errors point to two distinct issues: your custom MCP app expects OAuth-based Databricks credentials (not a raw bearer token), and the URL you pass to the HTTP transport must be a fully qualified https URL to the app’s /...

  • 0 kudos
Maylin
by Databricks Partner
  • 6749 Views
  • 1 replies
  • 0 kudos

Best Practices for Multilingual Model Training: Single vs. Multi-Model for Translation

Hello everyone,I’m working on a translation project involving documents up to 100 pages long, in 17 different languages, and I'm looking for the best approach to achieve high-quality translations in this multilingual context.Single model vs. multi-mo...

  • 6749 Views
  • 1 replies
  • 0 kudos
Latest Reply
Louis_Frolio
Databricks Employee
  • 0 kudos

Greetings @Maylin ,  A single many-to-many multilingual model, fine-tuned jointly across your 17 languages, is usually the best trade-off between quality, scalability, and operational simplicity; combine it with lightweight adapters to preserve quali...

  • 0 kudos
MarsWalker
by New Contributor II
  • 6914 Views
  • 2 replies
  • 1 kudos

How to serve a RAG chain endpoint that supports streaming

Hello everyone,I am trying to serve a sample RAG chain model that should support streaming output. But I could not find any documantation on how to enable streaming for a serving endpoint for a langchain model. Could you provide some hints on how to ...

  • 6914 Views
  • 2 replies
  • 1 kudos
Latest Reply
Louis_Frolio
Databricks Employee
  • 1 kudos

Greetings @MarsWalker ,  here’s how to get true streaming from a served RAG chain on Databricks Model Serving.   What’s going on The served endpoint error (“This endpoint does not support streaming”) is expected when the deployed entity doesn’t expos...

  • 1 kudos
1 More Replies
c4ndy
by New Contributor
  • 1256 Views
  • 1 replies
  • 0 kudos

Provisioned Throughput is not supported for llama...

Hi,This question is tightly correlated with another discussion: Model deprecation issue while serving on Databrick... - Databricks Community - 131968In a nutshell, I'm trying to serve the model which is based on llama architecture (deployed through m...

  • 1256 Views
  • 1 replies
  • 0 kudos
Latest Reply
Louis_Frolio
Databricks Employee
  • 0 kudos

Greetings @c4ndy ,  Thanks for sharing the context and links. I couldn’t read the Databricks Community thread from the URL you provided; please verify the URL and access settings, as it may not yet be available to Glean or require additional permissi...

  • 0 kudos
avadhut22111997
by New Contributor II
  • 6297 Views
  • 1 replies
  • 1 kudos

Request to share Genie space URL with restricted access

Dear Databricks Team,I hope this email finds you well.I would like to share a specific Genie Space URL with a user, but I want to ensure that the user only has access to view this particular Genie Space ask questions in that space and no other featur...

  • 6297 Views
  • 1 replies
  • 1 kudos
Latest Reply
Louis_Frolio
Databricks Employee
  • 1 kudos

Greetings @avadhut22111997 , sorry this fell through the cracks. I am Happy to help you lock down access to a single Genie Space so the user can only view it and ask questions there.   What “view and ask only” means in Genie Give the user the Genie S...

  • 1 kudos
tholers
by Databricks Partner
  • 11149 Views
  • 1 replies
  • 2 kudos

Data security when using Databricks Foundation Model endpoints

Hi Databricks,My client is considering using Databricks as the solution to their RAG chatbot framework. We are testing on the Databrick's demo tutorial on its own documentation data (pdf files). The tutorial is provided at: 00-RAG-chatbot-Introductio...

  • 11149 Views
  • 1 replies
  • 2 kudos
Latest Reply
Louis_Frolio
Databricks Employee
  • 2 kudos

Hey @tholers , here’s how Databricks handles data for the embedding model endpoints and what that means for tenant and data residency requirements. Short answer If you use Databricks-hosted embedding endpoints (for example, GTE-Large or BGE-Large), y...

  • 2 kudos
Khaja_Zaffer
by Esteemed Contributor
  • 1144 Views
  • 3 replies
  • 1 kudos

GEN AI IN HEALTHCARE

Hello everyone !!Good day!Happy halloween I am just curious to know those who are working with GEN AI or related projects in healthcare. Is GEN AI buzz word finally became a common word or still there is scope in real world projects. How do you this ...

  • 1144 Views
  • 3 replies
  • 1 kudos
Latest Reply
bianca_unifeye
Databricks MVP
  • 1 kudos

Hi @Khaja_Zaffer ,Great question  and happy belated Halloween From what we’re seeing across clients in healthcare and life sciences, Generative AI has absolutely moved beyond the buzzword stage. It’s now being operationalised, but selectively, in are...

  • 1 kudos
2 More Replies
ankitmishra
by Databricks Partner
  • 872 Views
  • 1 replies
  • 0 kudos

Resolved! RAG pipeline in Databricks

We have to implement a RAG pipeline in Databricks using Unity Catalog and Delta tables, We are a databricks partner company, can we have some credits to do over this PoC

  • 872 Views
  • 1 replies
  • 0 kudos
Latest Reply
bianca_unifeye
Databricks MVP
  • 0 kudos

Hi @ankitmishra,Yes, it is definitely possible to request credits for your PoC. The quickest route is to identify the Partner Solution Architect (PSA) assigned to your company account. They can submit the request on your behalf and even attach it to ...

  • 0 kudos
smferro54epam
by Databricks Partner
  • 1974 Views
  • 4 replies
  • 2 kudos

Can't connect to custom MCP Server

I've tried connecting from local environment, notebook and following the sample notebook here:https://docs.databricks.com/aws/en/generative-ai/mcp/custom-mcp?language=Local+environmentAnd I'm always getting the same error:httpx.ConnectError: All conn...

  • 1974 Views
  • 4 replies
  • 2 kudos
Latest Reply
smferro54epam
Databricks Partner
  • 2 kudos

I wasn't setting the MCP server as a streamable http, I've done that already and it works!

  • 2 kudos
3 More Replies