Tag
Anthropic redesigned Claude Code Projects into a coordinator-plus-parallel-worker architecture where each thread runs as an independent cloud session on its own Git branch, launching in public beta for select Claude Code subscribers.
Palomino 是一款 GPU 渲染的超快 AI 代理工作区,可在同一窗口中并行运行多个代理,并与终端、编辑器和浏览器协同工作。
Claude Code Projects 从窄范围 beta 正式向所有 Pro 和 Max 用户开放,用协调器替代文件夹式项目,可在并行云端会话中分配任务并共享项目记忆。
MemMux is a local runtime that treats resource governance for fleets of parallel coding agents as runtime verification, providing per-agent memory attribution, complete reclamation, escaped-process visibility, and bounded footprint. Benchmarks show it keeps a 7.5 GiB budget under zero swap and detects all injected escaped processes, with an open-sourced engine and benchmark.
Singularity is a Product Hunt launch that replaces long chat sessions with AI coding agents with atomic tickets, each scoped to an isolated Git worktree so agents can run in parallel across repos. It is free, local-first, and lets users bring their own API key.
GitHub Copilot app enables running multiple AI agents in parallel with isolated Git worktrees, allowing simultaneous tasks like building, reviewing, and testing to enhance developer productivity.
The paper introduces the 'stale' benchmark to study semantic coordination failures in parallel LLM-agent development, showing that interference is frequent in controlled tasks but rare in real-world reviewed pull requests.
The author describes an experiment where merging two AI agents' git worktrees led to test failures despite clean merges, highlighting the challenges of parallel agent development without mutual awareness.
Introduces `lane`, a CLI tool that uses copy-on-write to create efficient Git worktrees with durable in-project memories, preventing unnecessary rebuilds and supporting parallel development.
The release of pi-subagents introduces dynamic workflows and autonomous sub-agents to development environments, compatible with Claude Code for enhanced scriptable agent orchestration.
Cloudflare's security-audit-skill is a coding-agent tool that orchestrates parallel agents to audit codebases for vulnerabilities through a six-phase pipeline, producing structured reports.
Running 11 research agents in parallel for a data sweep cut processing time in half but revealed failures like empty briefs and API rate limits, emphasizing that orchestration and verification are more critical than the specific tool used.
The tweet argues that running too many AI coding agents in parallel degrades codebases and advocates a structured setup with a few specialized agents. It also quotes the launch of Jcode, an open-source agent claiming 20x memory efficiency.
A user reports that in Llama-CPP with parallel sub-agents, decode performance is great but a single agent's prefill (e.g., processing a web search) stalls all other agents, and asks for tuning suggestions.
1jehuang launched Jcode, an open-source terminal coding agent written in Rust that claims 20x better memory efficiency than Claude Code, allowing dozens of agents to run in parallel.
Multi-agent AI systems commonly fail at routing, parallelism, handoffs, and coverage. This post recommends a dispatch matrix, parallel execution, structured handoffs, and a catch-all fallback with logging to fix these issues.
Claude Code Merge Queue is a local, zero-cost merge queue that serializes landings from parallel Claude Code agents to avoid push races and build conflicts. It provides a CLI tool and configuration for managing parallel agent workflows.
Elon Musk announces a /deep-research command for Grok Build that performs research with bounded parallel agents, cross-checks evidence, and generates cited reports.
The article discusses three unexpected problems when running multiple coding agents in parallel: conflicts over shared working tree, runtime collisions (database, ports), and difficulty detecting stuck agents. Solutions include using per-agent git worktrees, isolated runtimes, and monitoring remaining gap metrics.