- 41429 Views
- 23 replies
- 51 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 ...
- 41429 Views
- 23 replies
- 51 kudos
- 12 Views
- 0 replies
- 0 kudos
Genie One Guardrails
HiWe are opening up the Genie AI to the business users. The Genie One seems to be a good fit, but the problem I noticed is the Guardrails and influence over the single orchestration. All the questions have been nicely interrupted by Genie one and wor...
- 12 Views
- 0 replies
- 0 kudos
- 103 Views
- 2 replies
- 0 kudos
Per-agent guardrails: what do you alert on, tokens, retries, or cost per task?
I'm trying to work out what a sensible guardrail looks like for each agent when several of them share a few serving endpoints.A plain token budget feels too blunt. An agent that retries a failing tool four times looks like a heavy user, while a cheap...
- 103 Views
- 2 replies
- 0 kudos
- 0 kudos
Short answer: I'd use all three, but at different layers. A per-agent rate limit is the hard backstop, a calls-per-request alert is the early warning for loops and retry storms, and cost per completed task against a 7-day baseline is a review metric,...
- 0 kudos
- 153 Views
- 3 replies
- 0 kudos
Databricks Genie One - Managing past conversation initiated by Databricks service principal account
Hello everyone.Recently, our team use Databricks service principal account, connecting to Genie One MCP server (via unity gateway) for data retrieval. However, we notice that there are currently no ways for us to manage the past conversations that ha...
- 153 Views
- 3 replies
- 0 kudos
- 0 kudos
Hi @ken_npk, Thanks for the detailed write-up. I think there is a supported way to list and delete service principal conversations, with one caveat about which MCP server you are using. The reason you can't find the service principal's conversations ...
- 0 kudos
- 259 Views
- 2 replies
- 1 kudos
Resolved! Monitoring a custom Agent (Responses) serving endpoint with an experiment on the Unity Catalog
Hi,I have a Langgraph agent deployed in Databricks on a custom serving endpoint of type Agent (Responses).So far, I have been tracing this agent on a "classic" monitoring experiment. I was using an ad hoc job to sync those traces to a table in the Un...
- 259 Views
- 2 replies
- 1 kudos
- 1 kudos
Hi! Thank you so much for your reply!So in the end, just adding MLFLOW_TRACKING_URI as env variables to the entity did the trick. Permissions were already set properly.
- 1 kudos
- 439 Views
- 5 replies
- 8 kudos
How do you regression-test a Genie space or agent when a table or metric view changes underneath it?
I hope this is useful for anyone running Genie spaces or agents past the demo stage. This is a question more than a recommendation, because I have not found a pattern I fully trust yet. The situation: a Genie space or an agent sits on top of a handfu...
- 439 Views
- 5 replies
- 8 kudos
- 8 kudos
Treat Genie like any other data product with a contract. Keep a golden set in Git, run it through the Genie Conversation API, compare result sets with tolerances rather than SQL text, and trigger runs from lineage and deploys. A silent shift is only ...
- 8 kudos
- 1259 Views
- 7 replies
- 10 kudos
Resolved! Genie Higher Environment deployment using DAB
Hi Team,I have successfully developed and tested the Genie in the development workspace environment. I now need to deploy the same Genie to higher environments, such as QA and Production, using the CI/CD process through DAB.Could someone please share...
- 1259 Views
- 7 replies
- 10 kudos
- 10 kudos
Thank you @ivanvyd, It is worked as expected!.And more thanks to @Ashwin_DSA!
- 10 kudos
- 569 Views
- 5 replies
- 0 kudos
Premium PAYG account stuck at TRIAL_VERIFIED (rate limit of 0 on resold models) — how do I get promo
Following up on the trust-tier mechanics iyashk-DB explained in this thread (https://community.databricks.com/t5/generative-ai/foundation-models-api-blocked-quot-databricks-set-rate-limit-of/m-p/160952) and this one (https://community.databricks.com/...
- 569 Views
- 5 replies
- 0 kudos
- 0 kudos
Short answer: I'm a community peer, but I don't have any information on the approval POCs either or I can't approve a tier change or push this into an internal queue. Nothing in the product or documentation gives an account like yours a self-service...
- 0 kudos
- 230 Views
- 1 replies
- 0 kudos
Unable to Serve Gemini 3.8 Flash on Azure Databricks – East US
I am working with an Azure Databricks workspace in the East US region and am trying to use the Gemini 3.8 Flash model through Databricks Model Serving.The model is visible in the workspace under the system AI models as:system.ai.databricks-gemini-3-8...
- 230 Views
- 1 replies
- 0 kudos
- 0 kudos
You're hitting that error because the create-endpoint flow is trying to build a provisioned throughput endpoint, and that isn't how you consume this model. system.ai.databricks-gemini-3-8-flash is a Databricks-hosted foundation model, and for propri...
- 0 kudos
- 884 Views
- 6 replies
- 5 kudos
Genie Agent refuses to add scalar SQL function from Unity Catalog
We would like to add some scalar SQL functions to our Genie Agents, for context that can't be provided via other mechanisms. In testing this, we have discovered that scalar functions are not supported, although the documentation explicitly states tha...
- 884 Views
- 6 replies
- 5 kudos
- 5 kudos
Greetings @charl-p-botha, I did some digging and here is what I found. This is a UI-path issue, not a limitation on scalar functions. The current docs say Genie supports both scalar and table-valued Unity Catalog functions, and the Aug 27 release not...
- 5 kudos
- 351 Views
- 2 replies
- 0 kudos
Resolved! Supervisor Agent Visualisations
I created a supervisor agent and asked it to plot something. It was able to generate text and table, then output a lot of gibberish which was supposed to be its attempt of plotting. Some of it's output (redacted):{'text/plain': '<Figure size 1400x100...
- 351 Views
- 2 replies
- 0 kudos
- 0 kudos
Thanks for the response.I tried your code and indeed the base-64 PNG rendered in a notebook, so it is a failure at the display layer. The agent is a Databricks-managed supervisor agent. I will create a support ticket and hope it gets looked at!
- 0 kudos
- 651 Views
- 3 replies
- 1 kudos
Routing between GLM 5.3 Flash and GLM 5.3 with ai_decide: a cheap router in front of ai_query
I hope this helps for those who are trying to keep LLM costs under control on batch workloads. I have been looking at ai_decide (Beta) as a router rather than as a classifier, and it fits that job well. Sharing the pattern and the open questions, sin...
- 651 Views
- 3 replies
- 1 kudos
- 1 kudos
On the three questions:1. Flash/full split at volume: I have not run ai_decide as a router at large batch volume either, so I would not invent a percentage. The number that matters is not a global flash share — it is flash share on the rows where a d...
- 1 kudos
- 333 Views
- 2 replies
- 0 kudos
Salesforce Hosted MCP + UC HTTP connection — stops working after ~1 hour?
We're wiring Salesforce Hosted MCP (read-only SObject server on api.salesforce.com) through a Unity Catalog HTTP connection and an MCP service, using OAuth User-to-Machine Per User.Login on the MCP service works and tools respond at first. After abou...
- 333 Views
- 2 replies
- 0 kudos
- 0 kudos
Your symptom points to the refresh path never firing. Once Databricks has an access token with no known expiry and no successful refresh, the per-user credential goes stale at the Salesforce-side lifetime and you get bounced to login. I'd treat the ...
- 0 kudos
- 381 Views
- 5 replies
- 0 kudos
What is the most effective way to evaluate Genie Agents/ Genie Spaces.
I would like to understand the most effective way to evaluate the Genie agent results. IS there any option we put as a guardrail for "Genie One" such that it doesn't go beyond the Genie spaces for searches apart from putting in prompt.
- 381 Views
- 5 replies
- 0 kudos
- 0 kudos
Short version: Genie One is a managed experience, so you don't get to pick the model or wire up the orchestration graph. What you control is the context and governance it works from, and that's where you'll get most of your leverage in a comparison....
- 0 kudos
- 2940 Views
- 4 replies
- 3 kudos
Resolved! Genie Code CLI support
Hi team,Currently the genie code is available only in the workspace UI. Do we have CLI capabilities for the genie code ? Or is it expected to come in the near future ? Please let me know your thoughts on this. Also, currently we have CLI support for ...
- 2940 Views
- 4 replies
- 3 kudos
- 3 kudos
Hey, we've just rolled out a Genie Code CLI to Beta. Check out the docs here: https://docs.databricks.com/aws/en/genie-code/genie-code-cliLet us know if you have any additional questions or feedback!
- 3 kudos
- 1014 Views
- 5 replies
- 4 kudos
Resolved! How do you test whether a retriever stops too early?
Hello everyone,Databricks’ Adaptive Instructed-Retriever announcement describes a model that learns when to stop searching. How would you test whether it stops after finding a general policy but before finding the exception that changes the answer?I...
- 1014 Views
- 5 replies
- 4 kudos
- 4 kudos
Hi @ivanvyd, sorry for the slow reply. Short honest answer: I have not run paired questions against the Adaptive Instructed-Retriever itself, so what follows is how I would set the test up, not results. Paired design, the way I would build it: take o...
- 4 kudos
