👾 New Coding Agents & Tools for Building Agentic Workflows

Welcome to Edition #7 of Agents Made Simple

There is a lot of noise in the AI coding space. Several new coding agents were released, promising to beat benchmarks and capabilities yet again.

Results often depend on the use case. I tested most of the coding tools and models, and the results were often different from benchmarks.

Whatever you’re trying to do with AI, think about testing several models side-by-side to determine the best fit.

This week’s topics:

• Google’s new coding agent invents algorithms

• OpenAI’s new coding agent capabilities

• Windsurf SWE-1 features and use cases

• Building agentic workflows with CrewAI

• Integrate an MCP flow with no code

• Plus AI investments, community highlights, trending AI tools, and more

AI Agent News Roundup

💥 Breakthroughs

📈 Investments

🇺🇸 Google launched the AI Futures Fund, a program designed to invest in startups. Startups can receive early access to advanced AI models, like Gemini, Imagen, and Veo, along with support from experts, cloud credits, and potential equity investment.

🇺🇸 Google is expanding Gemini. It will be available on more Android devices, including Wear OS smartwatches, Android Auto and cars, and Google TV, and integrated with the Android XR platform for future headsets and glasses.

🇺🇸 Google is developing a software AI agent. It’s designed to help software engineers with various tasks throughout the software development lifecycle, such as responding to tasks and documenting code.

🇺🇸 The Trump Administration’s Middle East visit secured new AI deals for Nvidia. Trump also published 3 image videos with 🇸🇦 Saudi Arabia, 🇶🇦 Qatar, and the 🇦🇪 UAE.

🇺🇸 Salesforce signed an agreement to acquire AI agent startup Convergence.ai to advance Agentforce. Convergence specializes in developing AI assistants capable of performing web tasks.

🇺🇸 YouTube announced new advertising solutions at Brandcast 2025, including Peak Points, an AI product using Gemini to place ads during highly engaging content moments.

🇺🇸 Netflix launched the Netflix Ads Suite, offering enhanced data, more buying options, expanded measurement, and creative ad formats that use genAI to integrate ads with their shows.

🇨🇳 TikTok introduced a new feature called AI Alive, available in TikTok Stories. It can transform static photos into dynamic, short-form videos.

🇨🇳 China launched the first 12 satellites for its “Three-Body Computing Constellation”. This project aims to build a network of 2,800 satellites capable of processing data using AI directly in space.

🇯🇵 Softbank wants to stall its $100B commitment towards OpenAI’s Stargate due to fears over U.S. tariffs and rising data center costs.

🇸🇦 Saudi Arabia launched a new state-owned AI company called Humain. It will focus on developing advanced AI models and building the necessary infrastructure.

Together with 1440 Media

Fact-based news without bias awaits. Make 1440 your choice today.

Overwhelmed by biased news? Cut through the clutter and get straight facts with your daily 1440 digest. From politics to sports, join millions who start their day informed.

Sign up now!

Overview of the CrewAI Framework for Building Agentic Workflows

1. What is CrewAI?

A Python-only (pip package) for building multi-agent systems with defined roles and workflows. Inspired by real-world team dynamics (roles, goals, tools). Designed for collaborative, multi-step AI agent systems. Ideal for beginners or projects needing quick setup for collaborative agents, with a growing but newer ecosystem.

2. Key Features

• Role-based agent definitions (like Engineer, Researcher, or Analyst).

• Sequential or parallel task planning using “Crew”.

• Integration with tools like browsers, search, APIs.

• Easy to customize with your LLMs and logic.

3. Benefits for Devs & Builders

• Rapid prototyping of complex multi-agent flows.

• Human-like collaboration logic (delegation, specialization).

• Lightweight yet powerful for most AI workflows.

• Open-source and actively maintained.

4. Real-World Use Cases

• An AI research crew that reads, summarizes, and fact-checks.

• Multi-agent customer support system.

• Automated report generation (e.g., stock analysis).

• More use cases here.

5. Tutorial

I’ve made a full hands-on tutorial where I build an AI sales offer crew step-by-step. Check it out on YouTube here.

6. What’s Next

In the coming issues, I’ll explore 5 more alternatives. Stay tuned for a full comparison!

Tool Spotlight

👾 Build a Dynamic MCP Flow You Can Integrate in Apps with Zapier MCP

Zapier offers a no-code platform to set up your individual MCP server. That gives you instant access to all 8000+ apps integrated in Zapier.

You just need to connect the apps you want to use and paste the MCP server configuration in your coding IDE like Cursor. Alternatively, you can integrate it into your agentic apps via Python or TypeScript.

🎛️ ACE-Step: Open-source foundation model for music generation, achieving SOTA performance, capable of running on consumer hardware.

📻 Stable Audio Open Small: Open-source text-to-audio model for generating music samples, capable of running offline on consumer devices.

Community Highlights

More Resources

Blog: AI-driven business automation and practical strategies for growth
AI Tool Collection: Discover and compare the perfect AI solutions
Consultancy: I help you solve your problem or discover AI potential