AI Radar

Your daily AI digest for developers — Saturday, March 21 2026

InfoQ AI

Stripe Engineers Deploy Minions, Autonomous Agents Producing Thousands of Pull Requests Weekly

Stripe engineers describe Minions, autonomous coding agents generating over 1,300 pull requests per week. Tasks can originate from Slack, bug reports, or feature requests, showcasing a practical implementation of agentic coding.

Why it matters: This highlights a real-world application of agentic coding, demonstrating how autonomous agents can significantly enhance productivity in software development.
InfoQ AI

Sonatype Launches Guide to Enhance Safety in AI-Assisted Code Generation

Sonatype Guide acts as a real-time guardrail system between AI coding tools and the open-source ecosystem, ensuring AI-generated code uses safe, valid, and secure components.

Why it matters: This guide helps developers mitigate security risks in AI-assisted coding, ensuring safer deployment of AI-generated code.
The Register AI

Sashiko: AI code review system for the Linux kernel spots bugs humans miss

Sashiko is an AI code review system designed for the Linux kernel, capable of spotting bugs that human reviewers might miss, thus improving code quality and reliability.

Why it matters: AI tools like Sashiko can enhance the accuracy and efficiency of code reviews, reducing human error and improving software quality.
Toward Data Science

Agentic RAG Failure Modes: Retrieval Thrash, Tool Storms, and Context Bloat (and How to Spot Them Early)

This article explores common failure modes in agentic RAG systems, such as retrieval thrash and context bloat, and provides strategies to detect and mitigate these issues early.

Why it matters: Understanding and addressing these failure modes can improve the reliability and efficiency of agentic coding systems.
MarkTechPost

A Coding Implementation Showcasing ClawTeam’s Multi-Agent Swarm Orchestration with OpenAI Function Calling

This tutorial presents the architecture of ClawTeam, an open-source Agent Swarm Intelligence framework, demonstrating how to orchestrate multi-agent systems using OpenAI's function calling.

Why it matters: Developers can leverage this framework to build complex, multi-agent systems capable of handling sophisticated tasks autonomously.
TechCrunch AI

WordPress.com now lets AI agents write and publish posts, and more

WordPress.com introduces AI agents capable of writing and publishing posts, potentially increasing the volume of machine-generated content on the web.

Why it matters: This development showcases the growing role of AI in content creation, automating tasks that traditionally required human input.
MarkTechPost

NVIDIA Releases Nemotron-Cascade 2: An Open 30B MoE with 3B Active Parameters, Delivering Better Reasoning and Strong Agentic Capabilities

NVIDIA's Nemotron-Cascade 2 is a new open-weight 30B Mixture-of-Experts model with enhanced reasoning and agentic capabilities, focusing on maximizing intelligence density.

Why it matters: This model provides developers with advanced tools for building more capable and efficient agentic systems.
Toward Data Science

The Math That’s Killing Your AI Agent

This article explains the compound probability math behind AI agent failures in multi-step tasks and introduces a pre-deployment framework to address these issues.

Why it matters: Understanding the mathematical underpinnings of agent failures can help developers design more robust AI systems.
TechCrunch AI

Microsoft rolls back some of its Copilot AI bloat on Windows

Microsoft is reducing Copilot entry points on Windows, starting with Photos, Widgets, Notepad, and other apps, to streamline user experience and reduce unnecessary complexity.

Why it matters: This move highlights the importance of balancing AI integration with user experience, ensuring AI tools remain helpful rather than intrusive.
Ars Technica AI

Widely used Trivy scanner compromised in ongoing supply-chain attack

The Trivy scanner, a popular tool for detecting vulnerabilities in container images, has been compromised in a supply-chain attack, highlighting the risks associated with AI-assisted security tools.

Why it matters: This incident underscores the need for robust security measures in AI-assisted development workflows to prevent supply-chain attacks.
✉ Subscribe to daily digest