Databricks Supervisor API adds background mode

// 45d agoPRODUCT UPDATE

Databricks Supervisor API adds background mode

Databricks’ beta Supervisor API lets developers hand off agent-loop orchestration, hosted tool execution, and final response synthesis to the platform. Background mode is the key piece: it keeps long-running, multi-tool workflows moving without blocking a synchronous request.

// ANALYSIS

This is Databricks moving from “host an LLM” toward “run the agent runtime for you.” The upside is less glue code and cleaner governance; the tradeoff is that you buy into Databricks’ hosted-tool model and its beta constraints.

–It removes the brittle part of agent apps: repeated model calls, tool selection, and result stitching
–Background mode makes it practical for workflows that query multiple systems or chain many steps
–The supported tool set is opinionated, which fits enterprise Databricks setups but narrows custom orchestration
–Limits matter: 30-minute max runtime, no client-side function tools, no exactly-once durability, no `temperature` control
–The cleanest fit is Databricks Apps plus Unity Catalog, Genie, MCP, and other governed internal tools

// TAGS

databrickssupervisor-apiapiagentautomationinfrastructuremcp

DISCOVERED

45d ago

2026-04-29

PUBLISHED

45d ago

2026-04-29

RELEVANCE

8/ 10

AUTHOR

OpenAI

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

MODEL52m ago

Zhipu AI drops GLM-5.2 coding model

Zhipu AI has launched GLM-5.2, its new 744-billion parameter Mixture-of-Experts coding model featuring a 1-million-token context window. Currently available to GLM Coding Plan subscribers, Zhipu AI plans to release the model's MIT-licensed weights and API next week.

MODEL1h ago

Claude Fable 5 exposes fragile safety guardrails

Anthropic's Claude Fable 5 was introduced as a safeguarded, public-facing variant of its powerful Mythos-class architecture designed for complex agentic workflows. The release strategy relies on real-time classifiers that reroute sensitive prompts to Claude Opus 4.8, depending on the premise that software guardrails can successfully isolate hazardous capabilities.

NEWS1h ago

Andrej Karpathy shares the "LLM Wiki" design pattern, where an AI agent actively maintains and organizes a user's knowledge base.

The "LLM Wiki" is a design pattern introduced by Andrej Karpathy to address the "memory rot" and organization challenges typical of second-brain systems. Instead of using standard Retrieval-Augmented Generation (RAG) to query a chaotic directory of notes, the pattern proposes a three-layer architecture: raw sources, a synthesized wiki of interlinked markdown files, and instructions for how the AI agent should maintain it. Under this pattern, the LLM acts as an active gardener of the wiki—synthesizing new info, identifying connections, and resolving contradictions—resulting in a compounding knowledge base.

Databricks Supervisor API adds background mode