읽을거리&정보공유

geeknews ,  meta-ai ,  yann-lecun ,  agi ,  facebook ,  world-model

google-colab ,  llm-finetuning ,  ocr ,  deepseek ,  deepseek-ocr ,  unsloth

chatgpt ,  openai ,  gpt-5-1 ,  gpt-5-1-thinking ,  gpt-5-1-instant ,  adaptive-conversational-experience ,  adaptive-reasoning

pdf ,  google ,  whitepaper ,  context-engineering ,  session ,  memory

serverless ,  ocr ,  pydantic ,  docx-to-markdown ,  tensorlake ,  document-to-json

code-agent ,  model-context-protocol ,  ai-code-reviewer ,  claude-code ,  mcp-server ,  claude-code-hook ,  quibbler

llm-in-production ,  llm-interpretability ,  any-llm ,  mozilla ,  mozilla-ai ,  any-llm-gateway ,  any-suite ,  any-agent ,  any-guardrail ,  mcpd

code-agent ,  agent-orchestration ,  multi-agent-framework ,  agor

pdf ,  dataset ,  ai-hub ,  guide ,  nia ,  training-dataset

cli-tools ,  gac ,  git-auto-commit ,  commit-message

nvidia ,  gdrcopy ,  gpudirect ,  gpudirect-rdma ,  cudamemcpy

safetensors ,  llm-inference ,  inference-framework ,  flashtensors ,  llm-hotswap

agent-framework ,  multi-agent-framework ,  agent-workflow ,  hephaestus ,  semi-structured-framework

pydantic ,  web-crawler ,  openrouter ,  hikugen

rag ,  pdf-gpt-indexer ,  local-rag

paper ,  calm ,  ai-ml-papers-of-the-week ,  ring-linear ,  continuous-autoregressive-language-models ,  llms-encode-how-difficult-problems-are ,  kosmos ,  ai-scientist-for-autonomous-discovery ,  train-for-truth-keep-the-skills ,  binary-retrieval-augmented-reward ,  graph-based-agent-planning ,  cache-to-cache ,  direct-semantic-communication ,  beyond-a-million-tokens ,  beam ,  from-memorization-to-reasoning ,  spectrum-of-loss-curvature

geeknews ,  llm-agent ,  from-scratch ,  context-engineering ,  agent-from-scratch

multi-llm-support ,  servicestack ,  open-webui ,  llms-py

geeknews ,  github ,  pull-request ,  ai-code-reviewer ,  claude-code ,  claude-code-action ,  github-action

파이토치

model-context-protocol ,  aws-mcp-proxy ,  mcp-proxy ,  sigv4 ,  amazon-q-developer

anthropic ,  claude-code ,  promotion

lsp ,  model-context-protocol ,  agent-client-protocol ,  acp ,  language-server-protocol ,  json-rpc

prompt-injection ,  meta-ai ,  llm-security ,  agents-rule-of-two ,  agent-security ,  oversight-agents ,  llamafirewall ,  human-in-the-loop ,  llama-protections

anthropic ,  awesome-list ,  voltagent ,  claude-skills ,  awesome-claude-skills

vscode ,  memory-framework ,  model-context-protocol ,  agent-memory ,  long-term-memory ,  llm-memory ,  openmemory ,  hierarchical-memory-decomposition ,  hierarchical-semantic-graph ,  memory-decay ,  explainable-recall-paths

geeknews ,  claude-code ,  vibe-coding

dataset ,  paper ,  survey-paper ,  large-reasoning-models ,  ai-ml-papers-of-the-week ,  vibe-coding ,  recode ,  deep-reasoning-dataset ,  sipit ,  real-deep-research ,  entropylong ,  dr-llm ,  lookahead-routing ,  fast-dllm-v2

geeknews ,  vibe-coding ,  addy-osmani