惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

酷 壳 – CoolShell
酷 壳 – CoolShell
H
Hacker News: Front Page
P
Palo Alto Networks Blog
T
ThreatConnect
Apple Machine Learning Research
Apple Machine Learning Research
博客园_首页
T
True Tiger Recordings
P
Privacy & Cybersecurity Law Blog
B
Blog
IT之家
IT之家
Last Week in AI
Last Week in AI
F
Full Disclosure
Hacker News: Ask HN
Hacker News: Ask HN
C
Comments on: Blog
Microsoft Azure Blog
Microsoft Azure Blog
C
Cybersecurity and Infrastructure Security Agency CISA
Microsoft Security Blog
Microsoft Security Blog
博客园 - 【当耐特】
N
News and Events Feed by Topic
NISL@THU
NISL@THU
腾讯CDC
雷峰网
雷峰网
Security Latest
Security Latest
李成银的技术随笔
M
Microsoft Research Blog - Microsoft Research
L
LangChain Blog
L
Lohrmann on Cybersecurity
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
C
Check Point Blog
Y
Y Combinator Blog
Recent Announcements
Recent Announcements
博客园 - Franky
N
News | PayPal Newsroom
V
V2EX
A
About on SuperTechFans
The Register - Security
The Register - Security
月光博客
月光博客
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Google Online Security Blog
Google Online Security Blog
MyScale Blog
MyScale Blog
Cisco Talos Blog
Cisco Talos Blog
Vercel News
Vercel News
WordPress大学
WordPress大学
C
Cyber Attacks, Cyber Crime and Cyber Security
The Hacker News
The Hacker News
IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
爱范儿
爱范儿
A
Arctic Wolf
L
LINUX DO - 最新话题
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More

Hacker News - Newest: "LLM"

PopuLoRA: Co-Evolving LLM Populations for Reasoning Self-⁠Play If an LLM is too expensive it won't be next year "This paper is LLM reviewed" > "this paper is peer-reviewed" StepStone: LLM-Based GPU Kernel Driver Fuzzing via User-Space Libraries [pdf] GitHub - AssimilatedHuman/LLM-Inquisitor: Evaluating AI behaviour under real‑world work conditions to surface issues before they become problems. LLM INQUISITOR identifies failures (drift, instability etc) by observing AI during normal tasks — a tool the industry desperately needs to stem the 85% failure rate. Includes Quick Start, Practitioner’s Guide and Methodology. Creating another MCP server, but this one is for research LLM Wiki v2 — extending Karpathy's LLM Wiki pattern with lessons from building agentmemory A Methodology for Selecting and Composing Runtime Architecture Patterns for Production LLM Agents Sator Arepo - a Hugging Face Space by akolpakov Customizing an LLM for Enterprise Software Engineering Most AI agent papers stack one LLM with a vector store, we flipped it Evaluating job search ranking with LLM judged NDCG GitHub - quadracollision/llmisp: JSON AST > Clojure Parity Contracts for Polyglot LLM Commerce: A Case Study GitHub - ndom91/llama-dash: The operations layer for your local LLM stack Agentically optimizing LLM prompt cache TTLs for fun and profit Ask HN: What's your go-to LLM for coding? How do you reduce LLM spam in PR reviews? Ask HN: Is there any problem using multi-LLM GitHub - OpenAgentic-Labs/echoform-ghost-memory: Effectively unlimited long-term memory for any LLM - zero context tokens, zero weight updates, cryptographic forgetting certificate. PSA — Posture Sequence Analysis Why More Context Can Make an LLM Worse GitHub - robertoranon/tokoro: A toolbox for building event publish & discovery web sites, apps, feeds, and more GitHub - sermakarevich/chunker: Agentic approach to chunking a document A new EDIT tool for LLM agents LLMCap — Hard Dollar Caps on LLM API Calls MLSys @ WukLab - Nitsum: Serving Tiered LLM Requests with Adaptive Tensor Parallelism SuperInfer: SLO-Aware Rotary Scheduling and Memory Management for LLM Inference on Superchips What political censorship looks like inside an LLM's weights — a mechanistic-interpretability study of Qwen 3.5 Managing metadata is essential in LLM world Fixing LLM Writing with Distribution Fine Tuning twitter.com Show HN: An LLM that's better at writing The local shape of LLM stable regions GitHub - msunda17/impactarbiter-cli The Infrastructure Behind Making Local LLM Agents Useful PostgreSQL ext makes LLM available as an index for similarity searches,inference GitHub - Tetrahedroned/Agent-Braille: Deterministic 8-bit machine-to-machine protocol for AI agent state. ~92% fewer state-tracking tokens on real Claude Code sessions, a proven single-bit-error-safe command code, fully reproducible. Tell HN: Writing an LLM critique/takedown? – Do not use an LLM to write it 🌱 an LLM models our worst behavior Prompt eval cues predicted refusal shifts across 32k LLM rollouts Ask HN: Is Java the ideal language for LLM-assisted coding? AI Foundry – Flat-Fee Unlimited LLM Inference on Blackwell GPUs in NZ LLM tracing with MLflow AI Gateway LLM Performance by Programming Language The LLM Looked Smart. The Metrics Disagreed – tiago.rio.br The Four Horsemen of the LLM Apocalypse GitHub - piqoni/piqo-extension: A good interface is invisible Intro to TLA+ for the LLM Era: Prompt Your Way to Victory Give every tool LLM wiki and bypass Claude Code SSH Throttle The Ultimate LLM Fine-Tuning Guide Ask HN: What LLM models are you using and why? Five Agents, One Browser: Werewolf on Quack + DuckDB LLM models are not ready for orchestrating many agents ClickBook — Offline AI eReader - Apps on Google Play DeepSeek-V4-Flash means LLM steering is interesting again Recent Developments in LLM Architectures: KV Sharing, mHC, and Compressed Attention Recent Developments in LLM Architectures: KV Sharing, MHC, Compressed Attention We Built SynapseKit: The Truth About Production LLM Frameworks GitHub - albedan/ai-ml-gpu-bench: A suite to benchmark CPU/GPU Python performance in training ML models and running local LLMs GitHub - chopratejas/headroom: Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server. if you are redlining the LLM, you aren't headlining Most Meaningful Dates on the Web and for an LLM I tested 8 LLM models on Linux without using the GPU RelaxAI – UK sovereign LLM inference at 80% cheaper than OpenAI/Claude GitHub - Andyyyy64/whichllm: Find the local LLM that actually runs — and performs best — on your hardware. Ranked by real, recency-aware benchmarks, not parameter count. One command, run it instantly. GitHub - krellixlabs/llm-reasoning-research: Curated, annotated research on reasoning gaps in large language models — temporal reasoning, causal reasoning, and beyond. Agentic evals or LLM as a judge? considering cost, time and quality Known By Their Actions: Fingerprinting LLM Browser Agents via UI Traces Add an LLM policy for `rust-lang/rust` by jyn514 · Pull Request #1040 · rust-lang/rust-forge GitHub - nimeshnayaju/markdown-parser: A streaming-capable markdown parser, written in TypeScript Dragos Documents First LLM-Assisted Strike on Water Infrastructure in Mexico Alchemize: PyMC's model to replace Stan/PyMC, etc. with an LLM BlitzGraph - The AI-native backend. Pokémon SVG Bench LLM Witch Hunts are getting F'in Irritating bliki: Interrogatory LLM Ctx-opt: TypeScript middleware to trim LLM chats to a token budget Show HN: Local-first Kubernetes YAML visualizer (no server, no LLM) Why Ruby Is the Better Language for LLM-Powered Development Paper page - Geometry Conflict: Explaining and Controlling Forgetting in LLM Continual Post-Training Show HN: Asciidia – LLM-Powered Game State media control shapes LLM behaviour by influencing training data Small Model Forensics How LLM Inference Works Multi-LLM AI trading agent harness GitHub - crawshaw/yeah: yeah: LLM-powered yes/no CLI tool Predicting Rare LLM Failures with 30× Fewer Rollouts — LessWrong Mechanism Design for Quality-Preserving LLM Advertising I tried to put an on-device LLM in an iOS Share Extension. It didn't fit Show HN: Gox – Strict static analyzer for Go designed for LLM-written code GitHub - torrix-ai/install Show HN: MCPSafe – Free security scanner for MCP servers using 5-LLM consensus Ada-MK: Adaptive MegaKernel Optimization via Automated DAG-based Search for LLM Inference Atlas Inference Engine Hi-Vis: one-shot jailbreak disguised as LLM "software patch" reaching 100% ASR Loading/running every LLM with 4M ctx in 3 clicks Free AI Leak Checker — Is Your Prompt Leaking Data? GLiGuard: 16x Faster Safety Moderation with a Small Language Model - Pioneer AI by Fastino Labs Are LLM Useful for Solo Founders
GitHub - genedeng-ca/ai-mac-migration: AI-powered Mac-to-Mac migration tool - replace Apple Migration Assistant with intelligent, selective transfer using local LLMs
2026-04-10 · via Hacker News - Newest: "LLM"

Install as Claude Code Skill

The fastest way to use this is as a Claude Code skill. One command, then just talk to Claude.

# Clone into your Claude Code skills directory
git clone https://github.com/genedeng-ca/ai-mac-migration.git ~/.claude/skills/mac-migration

Or if you just want the skill file without the full repo:

mkdir -p ~/.claude/skills/mac-migration/skill/scripts
curl -o ~/.claude/skills/mac-migration/skill/SKILL.md \
  https://raw.githubusercontent.com/genedeng-ca/ai-mac-migration/master/skill/SKILL.md
curl -o ~/.claude/skills/mac-migration/skill/scripts/scan.sh \
  https://raw.githubusercontent.com/genedeng-ca/ai-mac-migration/master/skill/scripts/scan.sh
curl -o ~/.claude/skills/mac-migration/skill/scripts/verify.sh \
  https://raw.githubusercontent.com/genedeng-ca/ai-mac-migration/master/skill/scripts/verify.sh
chmod +x ~/.claude/skills/mac-migration/skill/scripts/*.sh

Then tell Claude: "Help me migrate my Mac"

Claude will walk you through scanning your source Mac, building an intelligent migration plan, executing the transfer, and verifying everything landed correctly.

Standalone scripts

You can also use the helper scripts directly without Claude Code:

# Run on the source Mac to get a full system scan
./skill/scripts/scan.sh           # Human-readable report
./skill/scripts/scan.sh --json    # JSON output for automation

# Run on the target Mac after migration to verify
./skill/scripts/verify.sh user@source-mac.local

I haven't reinstalled macOS in 20 years. Every Mac I've owned has been migrated forward -- from PowerBook G4 to M5 MacBook Pro, across 8 machines, carrying two decades of muscle memory, scripts, configs, and digital life. Apple's Migration Assistant got me 80% there each time. AI got me the rest.


The Problem

Apple Migration Assistant is a black box. It copies everything -- the 50GB of Intel-only apps that will never run on Apple Silicon, the three conflicting Python installations, the orphaned preference files from software you uninstalled in 2019. You watch a progress bar for 6 hours and pray.

When it's done, you spend the next two weeks discovering what broke:

  • That app needs Rosetta and runs at half speed
  • Your SSH keys migrated but the permissions are wrong
  • 30GB of duplicate photos because iCloud and local copies both came over
  • VMware Fusion config migrated, but the actual VMs didn't
  • Software licenses tied to hardware IDs are now invalid
  • Homebrew packages compiled for Intel are silently failing

Migration Assistant treats your Mac like a disk image. Your Mac is not a disk image. It's a living system.

The Solution

AI Mac Migration replaces blind copying with intelligent transfer. It scans, understands, decides, and verifies -- using a 122-billion-parameter LLM running locally on your Mac. No cloud. No API calls. No telemetry. Just a smarter migration.

Source Mac ──SSH──> AI Analysis ──rsync──> Target Mac
                      │
               Local LLM on M5 Max
               Qwen3.5 122B via OpenClaw
               "Should this transfer?
                Is there a better version?
                Will this even work on ARM?"

How it works

  1. Scan -- Inventory both machines: apps, configs, packages, hardware capabilities
  2. Analyze -- LLM evaluates every item: architecture compatibility, version currency, size vs. value
  3. Plan -- Generate a ranked transfer plan with explanations for every decision
  4. Execute -- Adaptive rsync with real-time health monitoring
  5. Verify -- Post-transfer checksums and functional testing

Features

  • Intelligent app scanning -- Detects Intel vs ARM binaries, flags Rosetta-dependent apps, suggests native alternatives
  • Cross-machine package reuse -- If another machine on your network already has the ARM version of a package, grab it from there instead of downloading
  • Real-time hardware health detection -- Monitors disk I/O, CPU thermal, network throughput during transfer; backs off before your fans hit jet-engine mode
  • Adaptive transfer speed optimization -- Dynamically adjusts rsync parallelism and block size based on network conditions and disk speed
  • Smart exclusion lists -- AI-generated .migrationignore that understands what cache files, build artifacts, and temp data can be safely skipped
  • Selective config migration -- Migrates shell configs, SSH keys, git configs with permission fixup and path rewriting for the new machine
  • Automated photo dedup and organization -- Detect and eliminate duplicate photos across iCloud, local library, and manual backups before transfer
  • Software license migration -- Detect apps with hardware-locked licenses, generate a deactivation checklist for the old machine before wiping
  • End-to-end verification with checksums -- Post-migration integrity check comparing source and destination with per-file SHA-256 verification
  • Rollback snapshots -- APFS snapshot on target before migration starts; one-command rollback if anything goes wrong
  • Profile-based migration -- Pre-built profiles for "Developer", "Creative Pro", "Business" that prioritize what matters for each workflow

The Story

This tool was born from migrating an M4 Max MacBook Pro to an M5 Max MacBook Pro -- a machine carrying 20 years of accumulated digital life across 160TB of family data, dozens of development environments, and a fleet of AI services.

The migration was powered by Qwen3.5 122B running locally on the M5 Max itself via OpenClaw -- no cloud APIs, no subscriptions. The AI scanned both machines over SSH, made real-time decisions about what to transfer, and caught problems that Migration Assistant would have silently propagated.

What should have been a weekend of pain became an afternoon of supervised automation. The AI caught things a human would miss:

  • 12 Intel-only apps that had native ARM replacements available
  • 8GB of orphaned container images from a Docker version two releases old
  • SSH config referencing hosts that no longer exist
  • Homebrew packages that could be pulled from a local fleet machine instead of downloading 4GB over the internet

Tech Stack

Component Role
Python 3.12+ Core orchestration
Qwen3.5 122B via OpenClaw Decision engine running locally on M5 Max (no cloud dependency)
SSH Secure cross-machine communication
rsync Adaptive file transfer with resume support
APFS snapshots Pre-migration safety net
Homebrew Package reconciliation and cross-architecture detection

Quickstart

# On your NEW Mac:
git clone https://github.com/genedeng-ca/ai-mac-migration.git
cd ai-mac-migration
pip install -r requirements.txt

# Point it at your old Mac:
python migrate.py --source user@old-mac.local --target .

# Review the AI-generated migration plan before anything transfers:
# The tool will show you exactly what it wants to do and why.

Lessons Learned (The Hard Way)

Real migrations taught us what AI gets wrong. These are now guardrails in the tool:

Mistake What Happened The Fix
Photo loss AI classified "IMG_" prefixed files as iOS duplicates. Some were unique screenshots. Never delete -- only flag. Human reviews all dedup suggestions.
App folder miss Some apps store data in ~/Library/Application Support/ with a different name than the .app bundle. Map app bundles to their data directories using lsregister dump.
VMware forget Migration Assistant copied VMware Fusion but not the VMs in ~/Virtual Machines/. Explicit VM detection: scan for .vmx, .vmdk, .qcow2, Parallels bundles.
License gap Migrated apps that needed per-machine activation. Old machine was wiped. Licenses lost. Pre-migration license audit with deactivation reminders.
Permission drift SSH keys and GPG keyrings migrated with wrong permissions. Silent auth failures. Post-transfer permission validator: 700 for .ssh/, 600 for private keys.
Homebrew architecture mismatch Intel Homebrew (/usr/local) packages copied to ARM Mac. Binaries crash silently. Detect architecture, reinstall from source or pull from fleet ARM cache.

Architecture

ai-mac-migration/
├── migrate.py              # Entry point
├── scanner/
│   ├── apps.py             # Application inventory & architecture detection
│   ├── packages.py         # Homebrew/pip/npm package analysis
│   ├── configs.py          # Shell, SSH, git config discovery
│   └── hardware.py         # Hardware capability comparison
├── analyzer/
│   ├── llm_engine.py       # Local LLM interface (OpenClaw compatible)
│   ├── compatibility.py    # ARM/Intel compatibility checker
│   └── dedup.py            # File & photo deduplication engine
├── executor/
│   ├── transfer.py         # Adaptive rsync orchestration
│   ├── health_monitor.py   # Real-time system health during transfer
│   └── verify.py           # Post-transfer integrity checks
└── plans/
    └── migration_plan.json  # AI-generated, human-reviewed transfer plan

Contributing

This started as a personal tool for migrating my own Mac fleet. If you've felt the pain of Migration Assistant and want something better, contributions are welcome.

# Run tests
python -m pytest tests/

# Run a dry-run migration (no files transferred)
python migrate.py --source user@old-mac.local --target . --dry-run

License

MIT License. See LICENSE for details.


Built by someone who has mass-migrated too many Macs and finally got tired of doing it manually. Powered by Qwen3.5 122B running locally on Apple Silicon -- no cloud, no API keys, just your Mac understanding itself.