Multi-agent Systems

14 stories tagged Multi-agent Systems.

A Factory Built Its Own AI Brain Without Data Scientists

Rushabh Doshi built a 36-agent AI system to run Machinecraft's sales ops—no data team, no custom model training. Here's what he actually built, and what it can't tell you.

Three pixelated characters in a loop cycle with hypnotic eyes on the center character, illustrating repetitive AI…

Claude Code Loop Types: A Practical Breakdown

Dev Kapoor1 week ago

AI. Dev Kapoor1 week ago

Claude Code Loop Types: A Practical Breakdown

AI LABS maps five Claude Code agentic loops—from stateless to self-improving—and explains which use case each one is actually built for.

A presenter in a dark suit stands before a large screen displaying a glowing green and white humanoid robot head alongside…

Fusion Agents and Abacus AI Redraw the AI Attack Surface

Rachel "Rach" Kovacs4 weeks ago

AI. Rachel "Rach" Kovacs4 weeks ago

Fusion Agents and Abacus AI Redraw the AI Attack Surface

Fusion Agents and Abacus AI can now deploy live infrastructure on request. That's not just a productivity story—it's a security story worth understanding.

A man speaks passionately while discussing AI engineering concepts, with text overlay explaining how intelligence…

Multi-Agent Systems That Run for Days: Inside Factory's Missions

Yuki Okonkwo2 months ago

AI. Yuki Okonkwo2 months ago

Multi-Agent Systems That Run for Days: Inside Factory's Missions

Factory's Luke Alvoeiro says the bottleneck in software engineering is human attention, not AI intelligence. Here's how Missions tries to solve that.

Google's A2A Protocol Makes AI Agents Talk to Each Other

Rachel "Rach" Kovacs3 months ago

AI. Rachel "Rach" Kovacs3 months ago

Google's A2A Protocol Makes AI Agents Talk to Each Other

Google's A2A protocol standardizes how AI agents communicate across frameworks. LangSmith's new integration shows what interoperability looks like in practice.

A smiling person next to the Ultraplan app icon with a starburst symbol on an orange background

Claude Code's Ultra Plan: When Speed Meets Quality

Bob Reynolds3 months ago

AI. Bob Reynolds3 months ago

Claude Code's Ultra Plan: When Speed Meets Quality

Anthropic quietly released Ultra Plan for Claude Code. It uses parallel AI agents to plan projects faster—and execution follows suit. Here's what's happening.

AI Agent Design Patterns Raise New Regulatory Questions

Samira Barnes4 months ago

AI. Samira Barnes4 months ago

AI Agent Design Patterns Raise New Regulatory Questions

Google's new AI agent patterns—loop, coordinator, and agent-as-tool—demonstrate technical sophistication while surfacing unresolved compliance questions.

A man in a lab coat points at a glowing "BUILD" button surrounded by a crowd, with bold text asking about vibe coding's…

Vibe Coding Just Grew Up—And Nobody Knows What It Is Yet

Dev Kapoor4 months ago

AI. Dev Kapoor4 months ago

Vibe Coding Just Grew Up—And Nobody Knows What It Is Yet

Perplexity and Replit's latest releases show vibe coding evolving into multi-agent systems. But the real story is what we still don't understand.

Think Podcast bonus episode featuring four panelists discussing Opus 4.6 and GPT-5.3-Codex AI models

AI's Super Bowl: Claude Opus 4.6 vs GPT-5.3-Codex

Zara Chen5 months ago

AI. Zara Chen5 months ago

AI's Super Bowl: Claude Opus 4.6 vs GPT-5.3-Codex

Anthropic and OpenAI dropped major AI models within an hour. IBM experts break down what this coordinated chaos means for developers and enterprise AI.

A2A Overview presentation slide about building collaborative agentic systems, featuring Google Cloud branding and an agent…

Google's A2A Protocol: Standards for AI Agent Communication

Samira Barnes5 months ago

AI. Samira Barnes5 months ago

Google's A2A Protocol: Standards for AI Agent Communication

Google launches Agent2Agent protocol to standardize how AI agents communicate. Technical details, adoption questions, and what it means for multi-agent systems.

Neon glowing red "K5" text with dripping effect and yellow lightning bolt on dark textured background with "UPDATE IS…

100 AI Agents Working Together: Kimi's Latest Move

Zara Chen5 months ago

AI. Zara Chen5 months ago

100 AI Agents Working Together: Kimi's Latest Move

Moonshot AI's Kimi K2.5 deploys up to 100 parallel AI agents simultaneously. Here's what the benchmarks show—and what they don't tell you.

Laptop workspace with cables and tech setup, overlaid with "AGENT TEAMS" text and three colorful pixel art robot characters

Claude's Task System Changes How AI Agents Work Together

Bob Reynolds6 months ago

AI. Bob Reynolds6 months ago

Claude's Task System Changes How AI Agents Work Together

While flashier AI tools grab headlines, Claude's task orchestration system quietly enables something more practical: AI agents that actually coordinate.

A man gestures while explaining a graph showing performance declining as agent count increases from one to multiple agents.

Multi-Agent AI: Why More Isn't Always Better

Jonathan Park6 months ago

AI. Jonathan Park6 months ago

Multi-Agent AI: Why More Isn't Always Better

Exploring when multi-agent AI systems outperform single agents and when they falter.

Mastering Subagents: Key Design Decisions

Zara Chen6 months ago

AI. Zara Chen6 months ago

Mastering Subagents: Key Design Decisions

Explore the crucial design decisions in building effective multi-agent systems with subagents.