You Need to Learn This New Claude Function...

· Source: The AI Advantage · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Emerging Technologies & Innovation · Depth: Intermediate, long

Summary

This week's AI news highlights significant desktop application releases and a new text-to-speech model. Google introduced a free Gemini desktop app for Mac, which also integrates with NotebookLM, offering basic AI assistance and notebook access via a global hotkey. Anthropic launched a redesigned Claude desktop app with enhanced Claude Code features, allowing multiple simultaneous coding projects and integrated terminal views. Crucially, Claude introduced "Routines," enabling scheduled, cloud-based tasks like meeting summarization using connectors such as Zoom, which run without requiring the local machine to be active. Additionally, Google's Gemini 3.1 Flash text-to-speech model demonstrated advanced emotional range and expressiveness, surpassing existing solutions like Eleven Labs, and is available for free testing and API integration. A free, offline dictation app for Mac, powered by the Gamma 4 model, also emerged, offering local transcription capabilities.

Key takeaway

For CTOs and VPs of Engineering evaluating AI integration, the introduction of Claude's cloud-based "Routines" fundamentally changes how automated workflows can be deployed, eliminating the need for always-on local infrastructure. Your teams should explore this for scheduled tasks like data processing or content generation, as it offers a robust, machine-independent solution. Additionally, the advanced emotional capabilities of Gemini 3.1 Flash Text-to-Speech present new opportunities for enhancing user experience in applications requiring highly natural voice interaction.

Key insights

Cloud-based AI routines and advanced text-to-speech mark significant strides in AI application and accessibility.

Principles

Method

Claude's Routines allow users to define scheduled tasks with specific prompts, select models (e.g., Opus 4.7), set recurrence, and integrate connectors like Zoom for automated cloud execution.

In practice

Topics

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Engineer, MLOps Engineer, Consultant

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The AI Advantage.