Cursor & Windsurf Just Got a Chinese Rival

Today’s top AI Highlights:

1. AI Agent for Microsoft Office suite - better than Copilot

2. Alibaba’s agentic AI IDE competes with Cursor and Windsurf

3. DeepSeek quietly dropped V3.1 hybrid reasoning model

4. This AI Designer can design anything - logos, websites, interiors

5. Rube - Unified MCP server that connects you to 500+ apps

& so much more!

Read time: 3 mins

AI Tutorial

Building targeted B2B outreach campaigns is one of the most time-consuming aspects of sales and marketing. The challenge isn't just finding companies; it's discovering the right decision-makers, researching genuine insights, and crafting personalized messages that actually get responses.

In this tutorial, we'll build a multi-agent AI email outreach system using OpenAI GPT-5, Agno for orchestrating agents, and Exa AI for intelligent web search. This system automates the entire outreach pipeline - from company discovery to personalized email generation - delivering professional, research-backed outreach emails in minutes instead of hours.

Our multi-agent system conducts real research on each company using website content and Reddit discussions and ensures every email feels genuinely personalized.

We share hands-on tutorials like this every week, designed to help you stay ahead in the world of AI. If you're serious about leveling up your AI skills and staying ahead of the curve, subscribe now and be the first to access our latest tutorials.

Don’t forget to share this newsletter on your social channels and tag Unwind AI (X, LinkedIn, Threads) to support us!

Latest Developments

AI Agent for Microsoft Office Suite - Better than Copilot 💼 📊

Microsoft Office just became autonomous.

And this one’s way better than Microsoft’s own Copilot integration.

Shadow AI is an autonomous agent that handles entire Microsoft Office workflows from start to finish, turning a single command into a complete multi-step process across Excel, Word, PowerPoint, and Outlook.

Instead of switching between applications and manually transferring data, Shadow breaks down complex requests into executable steps and orchestrates everything behind the scenes. The agent maintains context throughout each workflow, ensuring that your Excel analysis flows seamlessly into your Word report, which then becomes the foundation for your PowerPoint presentation, all culminating in a polished email with attachments ready to send. It can execute complete business workflows with just simple prompts and no other intervention.

Key Highlights:

1. Cross-Application - Shadow seamlessly operates across Excel, Word, PowerPoint, and Outlook, using API calls and process automation to control multiple applications simultaneously.

2. Task Execution - The agent breaks down complex, multi-step requests into logical sequences of executable tasks, displaying its "chain of thought" reasoning process.

3. Context Persistence - Information and intent carry forward from one application to the next, ensuring Excel analysis directly informs Word reports, which then shape PowerPoint presentations.

4. Complete Workflow Automation - From data retrieval and analysis to document creation and email distribution, Shadow handles entire business processes without requiring manual intervention between steps.

Alibaba’s Agentic AI IDE Competes with Cursor and Windsurf 🖥️🤖

First came the LLM explosion; now Chinese AI companies are making their move on developer tools.

Alibaba just dropped Qoder, an agentic AI IDE that can think, plan, code, and execute to help you build real software and applications.

Think Cursor's interface with chat, agents, code completion, MCP integration, and other usual features, packaged with some more thoughtful additions for complex development workflows.

Here are the standout features: Quest Mode, where you can write specifications and let AI agents handle entire features while you grab coffee, and Repo Wiki that auto-documents your project while continuously tracking code and document changes. Qoder is heavily bent towards the concept of Context Engineering, where it deeply understands the codebase, design, and project architecture, maintains memory of past interactions, and automatically routes tasks to the most suitable models.

It’s 100% free during public preview, and a Pro tier will soon be introduced.

Key Highlights:

1. Spec-driven - Quest Mode transforms detailed specs into executable tasks, letting AI agents work autonomously on complex features while providing real-time progress tracking.

2. Codebase documentation - Repo Wiki automatically generates and maintains structured documentation that captures architecture decisions, dependencies, and technical context.

3. Adaptive model selection - Automatically chooses optimal LLMs based on task complexity without manual switching, ensuring efficient resource usage for different development scenarios.

4. Persistent project memory - Builds comprehensive memory of developer preferences, project patterns, and coding styles that improves AI suggestions over time.

Find out why 1M+ professionals read Superhuman AI daily.

AI won't take over the world. People who know how to use AI will.

Here's how to stay ahead with AI:

1. Sign up for Superhuman AI. The AI newsletter read by 1M+ pros.

2. Master AI tools, tutorials, and news in just 3 minutes a day.

3. Become 10X more productive using AI.

Join 1 million pros and start learning AI

Quick Bites

DeepSeek quietly dropped DeepSeek V3.1 hybrid reasoning model
DeepSeek released V3.1 this week, ditching the separate reasoning model approach for a unified hybrid model that switches between thinking and non-thinking modes. The model has gained significant improvements in agentic coding, reasoning, and multi-step tasks, comparable to Claude Sonnet 4 and GPT-5. Completely open-sourced under MIT license, available on DeepSeek chat, Hugging Face, and via API.

This AI Designer can design anything
Genspark AI has picked up a new skill: end-to-end brand creation. The new Genspark AI Designer can design literally anything - logos, packaging, websites, interiors, and everything else. It is a canvas-like experience where you can generate assets and iterate on them via prompts. Once it’s done, this entire visual workspace can be handed off to create a functional website, completing the loop from concept to code.

The Best Transcription Model for Prompting AIs
Aqua Voice, the AI dictation startup that lets you talk into literally any text field on Mac, just dropped Avalon, their new speech-to-text model that's specifically trained for how people actually talk to AIs. Think about those annoying transcription errors like "comment to get up" instead of "commit to GitHub." The model jumps from 85% to 96% accuracy on 500 common dev terms and beats OpenAI's Whisper Large V3 by 1.1% on industry benchmarks. Avalon rolls out to Aqua Voice Pro users this week.

Google’s AI Mode + Project Mariner for taking actions
Google’s AI Mode has gained huge agentic capabilities to help you get things done more easily. Starting with restaurant booking, the AI Mode can navigate multiple platforms and filter results based on your preferences for party size, cuisine, and timing. Under the hood, it uses the live web browsing capabilities of Project Mariner, partner websites, knowledge graphs, and Google Maps. This is the beginning towards AI Agents that go beyond search and take actions on your behalf.

Tools of the Trade

1. Rube: A unified MCP server that connects MCP clients like Claude and Cursor to 500+ apps like Gmail, Slack, GitHub, and Notion. Ask questions that involve actions across multiple apps - it auto-routes the queries to the right app in real-time without you manually instructing or configuring.

2. Claudable: Opensource Lovable that runs locally with Claude Code. Just describe your app idea and watch it generate the code with a live UI preview of your working app. You can deploy your app to Vercel and integrate databases with Supabase for free.

3. mcpd: A command-line tool that runs MCP servers from a config file. It handles the setup and management, so you don't need to manually start each server. It exposes all servers through a single HTTP API endpoint.

4. Awesome LLM Apps: A curated collection of LLM apps with RAG, AI Agents, multi-agent teams, MCP, voice agents, and more. The apps use models from OpenAI, Anthropic, Google, and open-source models like DeepSeek, Qwen, and Llama that you can run locally on your computer.
   (Accepting GitHub sponsorships now)

Hot Takes

1. If Jensen truly believed AGI was near, Nvidia wouldn't sell a single GPU. ~
   Yuchen Jin
   
   

2. your chatgpt history is your unfiltered cognition, the stuff you wouldn’t even tweet.

   your browsing history is your raw curiosity trail.

   your photos are frozen intimacy & accidental self portraiture.

   your messages are emotional telemetry.

   your mail is the transactional skeleton of your life.

   your calendar is the architecture of your past & future.

   your location history is where you actually go vs. where you say you go.

   your notes are a private brain dump, unpolished truths.

   once ai understands all of these at scale, it will unlock the most spiritual experience of your life. like an existential journey into the inner most you. ~
   signüll

That’s all for today! See you tomorrow with more such AI-filled content.

Don’t forget to share this newsletter on your social channels and tag Unwind AI to support us!

Unwind AI - X | LinkedIn | Threads

Awesome LLM Apps | Sponsor Us

PS: We curate this AI newsletter every day for FREE, your support is what keeps us going. If you find value in what you read, share it with at least one, two (or 20) of your friends 😉