Perplexity integrates Deep Research into Computer harness

// 2h agoPRODUCT UPDATE

Perplexity integrates Deep Research into Computer harness

Arav Srinivas announced that Deep Research is now a built-in, native skill within Perplexity Computer. Users no longer need to explicitly switch into a standalone mode for Deep Research; it simply works seamlessly as part of the agent harness, marking a significant step forward for state-of-the-art AI automation.

// ANALYSIS

This update is a huge quality-of-life improvement that makes complex, research-heavy workflows truly autonomous. By baking Deep Research directly into the Computer harness, Perplexity is blurring the lines between a traditional search engine and an end-to-end task execution agent.

–Eliminates context-switching friction for users.
–Connects in-depth research directly to multi-step execution.
–Strengthens Perplexity Computer's position as a unified AI orchestrator.

// TAGS

perplexity-computeraiagentdeep-researchautomation

DISCOVERED

2h ago

2026-06-11

PUBLISHED

2h ago

2026-06-11

RELEVANCE

9/ 10

AUTHOR

AravSrinivas

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

LAUNCH14m ago

Claude Fable 5 builds games with Higgsfield MCP

The Higgsfield MCP (Model Context Protocol) server enables Anthropic's Claude Fable 5 to connect directly to Higgsfield's suite of cinematic image and video generation tools. This connection allows the AI agent to autonomously manage the game development pipeline, producing storyline code alongside rich, customized visual assets without requiring the user to switch between separate media tools or browser tabs.

LAUNCH49m ago

Browser Use launches fetch-use scraping SDK

Browser Use has introduced fetch-use, a lightweight Python SDK and HTTP service that enables developers to scrape web content and retrieve search results without launching resource-heavy interactive browser sessions. Routing requests through stealth proxies to bypass bot detection, the tool supports session persistence and outputs markdown or structured JSON directly for LLM integration.

NEWS1h ago

Tesla FSD Supervised claims superhuman reaction times

Tesla has highlighted the safety performance of its FSD Supervised system, claiming its camera-only vision suite and neural networks achieve superhuman reaction times in safety-critical situations. While demonstrating significant safety-enhancing potential, users must still maintain constant supervision as the technology is under active development.