# Databricks AI agent for Slack & Microsoft Teams Viktor is an AI employee that connects to Databricks with a scoped API key and helps you run and evaluate LLM and AI infrastructure workloads, running 42 Databricks actions for you like Get Run, Get Job, and Reset Job. Ask in plain English in Slack or Microsoft Teams; Viktor does the work in Databricks and reports back. Setup takes about two minutes. - Tool: Databricks - Category: Other - Connection: API key - Actions available: 42 - Works from: Slack and Microsoft Teams - Page: https://viktor.com/integrations/databricks ## What can you ask Viktor to do in Databricks? - "Run this batch of documents through Databricks and summarize the results." - "Compare Databricks's output on this prompt with what we use today." - "Set up a daily job that queries Databricks on new entries and reports back." - "Watch Databricks for a cost or latency spike and alert me with the likely cause." ## What can Viktor do in Databricks? (42 actions) - **Upsert Vector Search Index Data** - Upserts (inserts/updates) data into an existing vector search index. - **Update Job** - Update an existing job. Only the fields that are provided will be updated. - **Sync Vector Search Index** - Synchronize a Delta Sync vector search index in Databricks. - **Stop SQL Warehouse** - Stops a SQL Warehouse by ID. - **Start SQL Warehouse** - Starts a SQL Warehouse by ID. - **Set SQL Warehouse Permissions** - Updates the permissions for a specific SQL Warehouse. - **Set SQL Warehouse Config** - Updates the global configuration for SQL Warehouses. - **Set Job Permissions** - Set permissions on a job. - **Scan Vector Search Index** - Scans a vector search index and returns entries after the given primary key. - **Run Job Now** - Run a job now and return the id of the triggered run. - **Reset Job** - Overwrite all settings for the given job. - **Repair Run** - Re-run one or more tasks. - **Query Vector Search Index** - Query a specific vector search index in Databricks. Returns up to Max Results items; when the result set spans more than one 1000-item page, the action automatically follows nextpagetoken to collect the rest. - **Query Vector Search Index Next Page** - Fetch the next page of results from a previously executed Databricks Vector Search query, using the nextpagetoken returned by a prior call. Run Query Vector Search Index first to get an initial page and its nextpagetoken, then pass that token here (and again on each subsequent call) to walk through all pages. Results are returned in groups of up to 1000, with a maximum of 10,000 total; an empty nextpagetoken in the response means there are no more pages. - **List Vector Search Indexes** - Lists all vector search indexes for a given endpoint. - **List SQL Warehouses** - Lists all SQL Warehouses available in the Databricks workspace. - **List Runs** - Lists all runs available to the user. - **List Jobs** - List all jobs using automatic pagination. - **List Endpoints** - List all vector search endpoints. - **Get Vector Search Index** - Retrieves details about a specific vector search index. - **Get SQL Warehouse** - Retrieves details for a specific SQL Warehouse. - **Get SQL Warehouse Permissions** - Retrieves the permissions for a specific SQL Warehouse. - **Get SQL Warehouse Config** - Retrieves the global configuration for SQL Warehouses. - **Get Run** - Retrieve the metadata of a run. - **Get Run Output** - Retrieve the output and metadata of a single task run. - **Get Job** - Retrieves the details for a single job. - **Get Job Permissions** - Get permissions of a job. - **Get Endpoint** - Get details of a specific vector search endpoint. - **Export Run** - Export and retrieve the job run task. - **Edit SQL Warehouse** - Edits the configuration of an existing SQL Warehouse. - **Delete Vector Search Index** - Deletes a vector search index in Databricks. - **Delete Data from Vector Search Index** - Deletes rows from a Direct Access vector index by primary-key values. - **Delete SQL Warehouse** - Deletes a SQL Warehouse by ID. - **Delete Run** - Delete a non-active run. Returns an error if the run is active. - **Delete Job** - Delete a job. Deleted jobs cannot be recovered. - **Delete Endpoint** - Delete a vector search endpoint. - **Create Vector Search Index** - Creates a new vector search index in Databricks. - **Create SQL Warehouse** - Creates a new SQL Warehouse in Databricks. - **Create Job** - Create a job. - **Create Endpoint** - Create a new vector search endpoint. - **Cancel Run** - Cancel a job run. The run is canceled asynchronously, so it may still be running when this request completes. - **Cancel All Runs** - Cancel all active runs for a job. The runs are canceled asynchronously, so it doesn't prevent new runs from being started. ## How does Viktor work with Databricks? 1. Add Viktor to Slack or Microsoft Teams. 2. Connect Databricks with an API key - Viktor handles authentication. 3. Ask in plain English. Viktor runs the Databricks actions, chains them with your other tools when needed, and reports back. Sensitive actions wait for your approval. ## What is a Databricks AI agent? A Databricks AI agent is an AI that connects to your Databricks account and completes work in it - it doesn't just answer questions about Databricks, it takes the actions. Viktor is that agent: an AI employee that works in Databricks on your behalf and delivers the finished result in Slack or Microsoft Teams. Unlike workflow builders such as Zapier or Make, there is nothing to configure - no triggers to map, no workflows to maintain. You describe the outcome in plain English, and Viktor picks the right Databricks actions, chains them with the other 3,200+ tools it connects to, and asks for approval before anything sensitive runs. ### About Databricks Databricks is the lakehouse company, helping data teams solve the world’s toughest problems. ## Security and permissions - Scoped API keys - Viktor stores your key securely, uses only the scopes you grant, and you can revoke it at any time. - Review-first approvals: sensitive Databricks actions wait for your sign-off before they run. - Isolated compute per team, encrypted in transit and at rest - and your data is never used to train models. - SOC 2 Type 1 compliant, with Type 2 and ISO 27001 in progress. ## FAQ ### Does Viktor integrate with Databricks? Yes. Databricks is one of the 3,200+ tools Viktor connects to, via a managed connector. Once connected, anyone on your team can put Viktor to work in Databricks from Slack or Microsoft Teams - no workflow builder, no code. ### How do I connect Databricks to Viktor? Ask Viktor in Slack or Microsoft Teams to connect Databricks, then paste an API key from your Databricks account when prompted. Viktor stores the key securely, uses only the scopes you grant, and you can revoke it at any time. Setup takes about two minutes. ### What can Viktor do in Databricks? Viktor can run 42 Databricks actions, including Get Run, Get Job, and Reset Job - and chain them with other tools in a single request. In practice that means Viktor helps you run and evaluate LLM and AI infrastructure workloads. If a task spans several actions, Viktor plans the sequence itself and reports back when the work is done. ### Do I need to build workflows to automate Databricks? No. Viktor isn't a workflow builder - there are no triggers to map and no workflows to maintain. You ask for the outcome in plain English in Slack or Microsoft Teams, and Viktor decides which Databricks actions to take, runs them, and reports back. ### Is my Databricks data secure with Viktor? Yes. Viktor connects to Databricks with a scoped API key, runs in an isolated environment per team, and never trains models on your data. Sensitive actions wait for your approval, and you can revoke access at any time. Viktor is SOC 2 Type 1 compliant, with Type 2 and ISO 27001 in progress. ### How much does the Databricks integration cost? Nothing extra - every integration, including Databricks, is available on every plan. You can start free with up to $100 in credits, no credit card required; paid plans start at $50/month. ### Can Viktor use Databricks together with my other tools? Yes - that's the point. Viktor connects to 3,200+ tools and works across them in a single run: it can pull data from Databricks, cross-reference it with another tool - for example Slack or Gmail, and deliver the finished result in Slack or Microsoft Teams. ## Viktor uses Databricks together with - [Slack](https://viktor.com/integrations/slack) - [Gmail](https://viktor.com/integrations/gmail) - [Google Sheets](https://viktor.com/integrations/google-sheets) - [HubSpot](https://viktor.com/integrations/hubspot) - [Notion](https://viktor.com/integrations/notion) - [Salesforce](https://viktor.com/integrations/salesforce) ## More Other integrations - [Front](https://viktor.com/integrations/front) - [BingX](https://viktor.com/integrations/bingx) - [TOPdesk](https://viktor.com/integrations/topdesk) - [Google Dialogflow](https://viktor.com/integrations/google-dialogflow) - [Google Slides](https://viktor.com/integrations/google-slides-2) - [Waitwhile](https://viktor.com/integrations/waitwhile) - [Dixa](https://viktor.com/integrations/dixa) - [Sendoso](https://viktor.com/integrations/sendoso) ## Get started Viktor is free to start - up to $100 in credits, no credit card required. All 3,200+ integrations are included on every plan. Sign up at https://viktor.com or browse all integrations at https://viktor.com/integrations.