Welcome to your weekly Brew & AI

Each week, I’ll share one blog, one tool, and one tip to make sense of AI - no jargon, no hype, just simple insights you can actually use.

P.S. - To ensure you receive this newsletter, please move it to the primary tab and set it for all future communications. Or just add this emailid as a known contact. I wish AI could solve this for me.

Grab your coffee - let’s dive in. 👇

☕ AI in the news

Back to another massive week of AI, updates all around

ChatGPT Groups: OpenAI has begun piloting Group Chats to GPT. You’ll be able to add people to a shared chat and upload files/create images together. Not only that, but you can react, reply and also see who’s typing. Collaboration in ChatGPT in real time. Looks like Slack and Teams won’t be around much longer.

Cursor Fundraise: AI coding tool Cursor is now worth $29.3B, (yes Billion). This line from the fundraise hits hard - “Our in-house models now generate more code than almost any other LLMs in the world” - all while competing against ChatGPT Codex and Claude Code. Would definitely recommend giving Cursor a go. I’ll be sharing a tutorial on how to set it up soon, stay tuned.

Google Shopping: Google Shopping has launched new AI-powered tools for holiday shopping, including conversational search, local inventory checks, and “agentic checkout” that enables Google to buy tracked items for you when prices drop. These innovations aim to make shopping faster, smarter, and more convenient this holiday season.

☕ Tool of the week

Replit: This week, we’re covering Replit, Lovable’s slightly more technical cousin.

In essence, Replit allows you to do all Lovable does, which is prototype super fast, but it offers additional features for those who want to get deeper into the technicalities.

Replit is a more code-focused platform built for developers and students who want to write, iterate, and deploy real functional applications in the browser. It supports many programming languages and allows you to build more stable, full-stack apps with coding and hosting capabilities. It is better for longer-term and production-ready projects, but requires more technical skills.

All other tools here

☕ Tip of the week

This one’s a gem. I’ve been using it to help me learn n8n (workflow building) and also to build a pipeline to generate tweets for me daily.

You want to learn AI. The problem is that each day brings a new tool, a new feature, and progress in AI moves at the speed of light. You're stuck with a bunch of feature releases, tutorials and no clue how to progress.

Use AI to learn AI. Chat with your favorite LLM, ask it to explain what each tool does, how it fits into your daily workflow, and how to build with it. Explicitly call out your comfort level, and ask it to walk you through every feature in detail. Then give it a use case and ask it to walk you through how to build using any tool.

  • For beginners, explicitly mention you know nothing, and ask it to explain to a toddler (This works wonders)

  • Any issues you encounter, just screenshot or copy it back into the chat to receive an immediate solution.

Pro tip: Follow the blog on how to write a good prompt, and be as clear as you can, don't feel shy to add any details you want.

You are an expert n8n workflow automation engineer with 5+ years of experience building production-grade AI agents. Your task is to help me build a complete n8n AI agent workflow for generating Twitter posts to help build my brand.

Use case: I want to build a content pipeline in line with my brand to help me post daily on twitter to gather engagement. I wish to post across themes such as AI news, AI tools, AI simplified, practical AI tips. The pipeline should run weekly and generate 10-20 posts per theme.
Data sources: RSS feeds, excel sheets, product Hunt
Desired outputs: The final output should be a list of tweets, applicable hashtags and a feedback mechanism where I can add a thumbs up/down for the posts and that should help build better tweets. 

Complexity level: Beginner

REQUIREMENTS:
1. Design the complete workflow architecture
2. Provide step-by-step n8n node configuration
3. Include error handling and retry logic
4. Add data validation and transformation steps

If you need any other information ask me

More tips here

☕ This week’s blog

Prompts Explained: How to Talk to AI (Without Feeling Awkward) - Brew & AI

Talking to AI doesn't have to feel weird. Learn how prompts work and why they're simpler than you think, no tech degree required. Also, get a clear blueprint on what the best prompts look like, and how to drastically improve the quality of your conversations with LLMs.

brewandai.com/blog/prompts-explained-how-to-talk-to-ai-without-feeling-awkward

💛 P.S.

That’s it for this week’s brew. We’ll soon be moving to tutorials and more practical content, I’m also exploring converting the blogs to short form video content, so stay tuned.

I’d love to hear what you think - what you liked, what could be better, or what you’d love to see next.

Just hit reply - I read every message over my morning coffee ☕.

Brew & AI
Making AI simple, one sip at a time

Keep Reading