Agentic Cloud Environment to Build & Deploy AI Apps

Today’s top AI Highlights:

1. Google’s agentic environment to build, test, and deploy full-stack apps

2. Fully-managed RAG platform with just 3 API calls

3. Grok-3 API with $150 in free API credits each month for enrolling in data sharing

4. China’s Moonshot AI released small but very powerful VLMs

5. Turn your FastAPI endpoints into MCP servers in a few lines of code

& so much more!

Read time: 3 mins

AI Tutorial

While tech bros are busy building autonomous AI agents to optimize your shopping or debug your code, let's focus on something that actually matters: healing your broken heart. Breakups are universally painful, and sometimes what you need isn't another self-help book or a friend who's tired of hearing about your ex—it's an emotionally intelligent AI system that won't judge you for stalking your ex's Instagram at 3 AM.

In this tutorial, we'll build a multi-agent AI breakup recovery application using Streamlit and Agno, powered by Gemini 2.0 Flash. This team of specialized AI agents will provide emotional support, help you craft cathartic unsent messages, plan recovery routines, and even give you some brutal honesty when needed.

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, Facebook) to support us!

Latest Developments

Build Full-Stack Production-Ready Apps with NoCode 🪄📲

Google has released Firebase Studio, an agentic cloud-based development environment that helps you build and ship production-quality full-stack AI apps, including APIs, backends, frontends, mobile, and more. The platform unifies Google’s AI IDE Project IDX with specialized AI agents and assistance from Gemini to provide a collaborative workspace accessible from anywhere.

You can either create apps through natural language prompts or write code in a familiar IDE environment, giving you flexibility based on your preferences and skills.. During the preview period, three free workspaces are available, with Google Developer Program members getting access to more.

Key Highlights:

1. Build Apps With or Without Code - Create apps by writing code or using the App Prototyping agent that turns your descriptions into working software. Generate entire Next.js web apps through natural language, images, or drawings, with AI features already configured and ready to use.

2. Customizable Development Environment - Import projects from GitHub, GitLab, or Bitbucket, or choose from 60+ templates for popular frameworks. The platform runs on a full virtual machine you can customize with your preferred tools, packages, and configurations.

3. AI Coding Support - Get help from Gemini for writing code, fixing bugs, running tests, managing dependencies, and creating documentation. The AI understands your entire codebase context and can implement new features through conversation.

4. Testing and Deployment - Preview web and Android apps directly in the browser, test with built-in emulators for Firebase services, and deploy to Firebase App Hosting with a single click. Share your work instantly and collaborate in real-time with team members.

RAG Made As Easy As Stripe Checkout 🤌📚🔍

LiquidIndex brings RAG capabilities to your applications with a surprisingly simple API that requires just three calls to go from zero to production.

Just like Stripe Checkout simplified payments, LiquidIndex handles all the complex RAG infrastructure—data processing, vector storage, and retrieval mechanics—behind clean API endpoints that developers can implement in minutes. The multi-tenant architecture isolates each user's data automatically, making it perfect for SaaS applications where customers need their own private document search.

Key Highlights:

1. Frictionless Integration - Add AI-powered document search to your application with just three API calls: create a customer, start an upload session, and query. The hosted upload interface handles document processing and embedding generation.

2. Built for Multi-Tenant Apps - Each of your users gets their own isolated data space with automatic query routing. Your application can scale to thousands of users without managing complex vector database partitioning or dealing with data separation concerns.

3. Complete Infrastructure Management - LiquidIndex handles the entire backend stack including file parsing, chunking, embedding generation, and vector storage. This removes the typical headaches of connecting and maintaining multiple systems like document processors, embedding models, and vector databases.

4. Data Sources - Currently supports direct file uploads and Google Drive integration, with Notion, Dropbox, and S3 connections coming soon. You can also customize vector databases and embedding models to avoid vendor lock-in.

Quick Bites

Google released the Live API for Gemini in Preview, allowing developers to build applications that process streaming audio, video, and text with minimal latency. The update includes support for 30 additional languages with new voice options and the ability to handle virtually infinite sessions through a sliding context window.

Cloudflare has launched Realtime, a suite of products for building interactive apps with real-time audio and video capabilities. It includes RealtimeKit, a developer platform that handles the WebRTC and streaming infrastructure with near-zero latency network worldwide, and provides SDKs for seamless client-side integration.

Google’s video generation model Veo 2 is now available via the Gemini API for developers with Tier 1 billing and above. The service supports both text-to-video and image-to-video generation, creating 720p videos at 24fps with lengths between 5-8 seconds in both landscape and portrait orientations.

xAI has released its flagship Grok-3 model via API in two versions. The standard Grok-3 costs $3 per million input tokens and $15 per million output tokens, the same as Claude 3.7 Sonnet. Another version, Grok 3 Mini (with Thinking), is a lightweight model that thinks before responding, costing $0.3 per million input tokens and $0.5 per million output tokens. You can get $150 in free API credits each month for enrolling in data sharing. Elon Musk has also confirmed that Grok-3.5 is releasing soon and Grok-4 is coming later this year.

Memory in ChatGPT can now reference all of your past chats to provide more personalized responses, drawing on your preferences and interests to make it even more helpful. You can opt out of referencing past chats or memory altogether, at any time in settings.

China’s Moonshot AI released Kimi-VL, a lightweight mixture-of-experts vision language model that gives advanced multimodal reasoning, long-context understanding, and strong agent capabilities—all while activating only 2.8B parameters. The model outperforms much larger flagship models like GPT-4o-mini, Gemma 3 12B, and DeepSeek VL 2 across all vision and agent benchmarks.

Building on this model, Moonshot AI also released an advanced long-thinking variant: Kimi-VL-Thinking with chain-of-thoughts reasoning that offers strong long-horizon reasoning capabilities. Both models are available to download on Hugging Face.

Tools of the Trade

1. Haystack Editor: Turns complex diffs into structured sections for visual review. It displays pull requests on an infinite canvas using AI to organize. It supports multiple programming languages and includes collaboration features directly on the canvas.

2. Browserable: Opensource browser automation library that enables AI agents to perform web tasks, track their progress, and retrieve results. It currently achieves 90.4% on the WebVoyager benchmark.

3. FastAPI_MCP: Turn your existing FastAPI endpoints into MCP tools with just a few lines of code. No need to rewrite your API or learn a new framework. It automatically discovers all FastAPI endpoints and converts them to MCP tools.

4. Awesome LLM Apps: Build awesome LLM apps with RAG, AI agents, and more to interact with data sources like GitHub, Gmail, PDFs, and YouTube videos, and automate complex work.

Hot Takes

1. It's hard for Europeans to comprehend how much money US developers make. Like, really, really hard. ~
   Santiago

   
   

2. Anthropic pushes MCP. Google rushes A2A. Welcome to the AI Standard Wars.
   "But they are complementary! Each standard has a purpose"
   Oh my sweet summer child. This is war. ~
   Solomon Hykes

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 | Facebook

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 😉