MinDong @petanerd
AI Security, Time-series prediction Joined July 2014-
Tweets121
-
Followers30
-
Following368
-
Likes2K
As an AI Infrastructure Engineer Please learn: - GPU architecture, VRAM fundamentals, CUDA & memory hierarchy - Quantization (INT8/FP8/4-bit), batching & continuous batching - vLLM, TensorRT-LLM, SGLang, llama.cpp & inference optimization - KV caching, speculative decoding, prefix caching & token throughput - Distributed training (DDP, FSDP, DeepSpeed, ZeRO) - Model serving (Triton, vLLM, KServe, Ray Serve, SGLang) - Kubernetes, Docker & GPU orchestration - NCCL, InfiniBand & high-speed networking - Multi-GPU & multi-node inference - LoRA, QLoRA, PEFT & fine-tuning pipelines - Vector databases, embeddings & RAG pipelines - Prompt caching, semantic caching & cost optimization - Observability (OpenTelemetry, Prometheus, Grafana, Langfuse, Phoenix) - LLM evaluation, benchmarking & A/B testing - Model routing & fallback strategies - MCP, AI agents & workflow orchestration - Data pipelines, Kafka & streaming inference - Security, guardrails & prompt injection mitigation - CI/CD for ML (MLOps), MLflow & model registries - Linux, networking & storage fundamentals
@PyTorchKR #토치야호 #torchyaho #파이토치데이 #2026 #PyTorchDay2026 #PyTorchDay #파이토치 #PyTorch #PyTorchKR
PyTorch Day Korea 2026! 올해 슬로건과 일자를 공개합니다! 🌟 “Torch YAHO!” 🌟 2026.10.31 - 센터필드 이스트 18층 AWS 코리아 교육장 PyTorch를 사용하는 연구자, 엔지니어, 개발자들이 모여 모델 개발부터 최적화, 서빙, AI Application까지 이어지는 다양한 경험을 함께 나눕니다. (1/3)
좋은 연구주제 잡는 법: 0. 우선 연구 목적을 명확히 한다. 논문만 나가면 되는지 (이때 어느 레벨로 논문이 나가야하는지), 공부만 하면 되는지, 자기만족인지, 진짜 연구인지 순으로 나뉜다. 순서가 이상해보일 수 있지만, 이 순서가 맞다는 게 (핵심1)이다. 우선 방법 1번부터 타래에 이어 씀.
학부연구생 2년차면 아직 3~4학년이라는 뜻인데... 좋은 연구 주제 잡기는 박사과정 3~4년차에게도 어렵습니다. 이건 사실 지도교수님이 좀 더 신경써주셔야 하는 부분이 맞습니다. 그런데 아이디어라는건 보통 실패한 실험에서 ... spin-spin.com/q/6a3132247861…
순천만 국가정원 스탬프찍으러 왔는데 정말 너무너무 좋았음!
software engineers before vs after agents
AI agents without guardrails aren't powerful, but they're risky. 9 layers every team needs before going to production... AI agents are only as reliable as the guardrails behind them. This is no longer optional; it's the difference between scaling safely or failing fast. Here are the essential guardrails every AI agent needs today: 1/ Content Filtering: Input + Output Stops harmful, sensitive, or non-compliant data before it enters or leaves your system. 2/ Input Validation: Query Stage Prevents prompt injection, enforces schema rules, and ensures structured inputs reach the agent clean. 3/ Intent Recognition Understands what the user actually wants, critical for correct tool routing and planning decisions. 4/ Rule-Based Checks: Pre-Processing Lightweight filters (regex, limits, constraints) that catch edge cases before reasoning even starts. 5/ Hallucination Detection: SLMs + Evaluators Flags low-confidence or fabricated outputs before they ever reach a user. 6/ Safety Classification: Specialized Models Classifies queries in real-time to block unsafe or restricted actions at the gate. 7/ Moderation Layers: APIs + Internal Models Adds redundancy across input and output because one layer is never enough in production. 8/ Output + Format Validation Ensures responses are usable (JSON, SQL, API-ready) and won't break downstream systems. 9/ Sensitive Data Detection: PII + Secrets Prevents leakage during both retrieval and generation. Non-negotiable for any enterprise RAG pipeline. 📌 What's shifted from 2025 to 2026: * Guardrails are now multi-layered systems, not single filters * Real-time evaluators and agent monitoring frameworks are standard * Policy-aware agents with compliance baked into logic, not bolted on * SLMs handling safety tasks faster, cheaper, purpose-built * "Defense-in-depth" is the architecture pattern enterprises are adopting Save 💾 ➞ React 👍 ➞ Share ♻️ & follow for everything related to AI Agents.
Minha prima é crocheteira . Pedi pra ela fazer ,um descanso de copo pra colocar na mesa do computador, Ela me vem com esse kkkkk
Ever suspected a paper you’re reading is AI slop? You can now turn on AI detection mode on alphaXiv to visualize what is written by an AI and what is not. Now available for every research paper indexed on arXiv. Integrated with the latest detection model from @pangram 🚀
@jun_song 안녕하세요. 원본 26b나 31b에는 오디오 모달리티가 없던데 혹시 파인튜닝으로 해당 모달리티를 추가하고 정렬하실 계획도 있으실까요?
This is Algebrica. A mathematical knowledge base I’ve been building for 2.5 years. 215+ entries, carefully written and structured. 400k+ views over this time. Not much in absolute terms, but meaningful to me. No ads. No courses to sell. No gamification. No distractions. Just essential pages, aiming to explain mathematics as clearly as possible, for a university-level audience. Built simply for the pleasure of sharing knowledge. Content licensed under Creative Commons (BY-NC). Best experienced on desktop. If it helps even a few people understand something better, it’s worth it.
#이벤트 11월 11일 빼빼로데이? ‼️🍟K-MEX 김치 프라이즈 데이 🍟‼️ RT 하면 K-MEX 김치 프라이즈가 찾아갑니다 😎 @cucharakorea 팔로우 + #RT 🎁 5명 당첨! 📆 이벤트 기간: 11.11 ~ 11.17 (당발 11.19) 지금이 바로, 김치프라이즈 타이밍 😉 #쿠차라 #Cuchara #Kmex김치프라이즈 #신메뉴 #RTevent #빼빼로데이대신김치프라이즈 #멕시칸푸드
@cucharakorea 오픈 첫날 첫끼는 부리또볼 스테이크+ 까르니따스 더블고기에 풀토핑 +과카몰 추가로 풍부하게 누리기! 카드지갑과 타바스코소스 미니어쳐 감사합니다:)
@cucharakorea 쿠차라 상품권을 받았는데 다음주에 회사앞에 쿠차라가 오픈을 한다? 이건 운명이라고 봅니다..😻 #cuchara #쿠차라 #부리또볼 #멕시칸푸드
프롬프트에 악의적 의도는 AI 브라우저에서도 작동될 수 있음. AI 어시스턴트는 스크린샷을 찍고 이걸 사용자에게 분석해줌. 문제는 스크린샷을 찍은 웹페이지에 숨겨진 지침도 따른다는 것. 여기서 문제가 생기는데 AI가 사용자의 인증된 권한으로 행동할 수 있다는 것.
로컬 LLM의 보안 역설 - 로컬 환경에서 프라이버시 보호를 위해 LLM을 사용하는 개발자들이 오히려 보안 취약성에 노출될 수 있다는 연구 결과가 제시됨 - 연구팀은 `gpt-oss-20b` 모델을 대상으로 한 OpenAI Red‑Teaming Challenge 실험에서, 로컬 모델이 공격자의 프… news.hada.io/topic?id=23862
Jinsoo Heo @devkoriel
1K Followers 452 Following Sr. DevOps Engineer @ChronicleLabs 🇪🇺🇺🇸 | ex-@Official_Upbit ex-@daangn_official ex-@PUBG | Kubernetes, GitOps, eBPF | Oracle infra keeping DeFi alive
seungho kim @raccoonyy
2K Followers 620 Following 다양함이 어우러지는 세상을 꿈꾸는 Software Engineer. 44bits, Write The Docs Seoul 운영진. 인프런에서 도커 컴포즈를 배워보세요: https://t.co/rTe9INzG4Z
lucas @lucas_flatwhite
15K Followers 4K Following ai x software engineer / art enthusiast / capitalism x investment / cat in pic https://t.co/N61jJXagbK
Aarn Crissy @jerrie84258
2 Followers 92 Following Recruiting webshell engineers to penetrate websites, with a monthly salary of up to $ 100,000. If interested, please contact https://t.co/P5P8o030tv
fafa.👩🏻💻 @ds_fafa_
29K Followers 10K Following Data Maven with a Dash of Espresso ☕️ | Turning Numbers into Narratives | Senior Customer Insights Director | Tweets fueled by caffeine and curiosity
포코 @yurisalgu
2K Followers 190 Following
MirabelleMond @arQTWP7y80Y35
71 Followers 1K Following
Mattou @Mattou448017
7 Followers 632 Following After walking for a long time, I realized that the distance is still full of tea, rice, oil and salt.
Sangmin Bae @raymin0223
1K Followers 1K Following Postdoc @kaist_ai | PhD @kaist_ai; Prev-Intern @Meta @GoogleDeepMind, Kakao
david gary birchmire ... @CrossfitBirch
2K Followers 5K Following https://t.co/4nA96F1qW8 https://t.co/f90AgyDVfm https://t.co/QPm63jqEhg
마이크로소프트 @MicrosoftKorea
5K Followers 3K Following Microsoft 목표와 가치는 전세계의 사람과 기업이 잠재력을 최대한 발휘할 수 있도록 돕는 것입니다.
Seoyeon Stella Yang @codestella
681 Followers 750 Following AI Roboticist Girl 🙌💕 3d Vision / Diffusion / 3d generation / NeRF / Visual localization / Vision AI Phd candidate in Seoul National University @SNUnow👩💻
북구 미래 하정�... @JungWooHa2
11K Followers 3K Following 전) 청와대 AI미래기획수석비서관 전) 네이버클라우드 AI 혁신 센터장 한국공학한림원 정회원 Ex) Sr. Secretary to the President for AI & Future Planning, #Korea
Better Proposals @betterproposals
15K Followers 12K Following Sales Proposals Made Simple. Try it free today.
Brunda Ganja @_avseq
2K Followers 5K Following
Chris Ohk @utilforever
4K Followers 1K Following Visiting Senior Researcher at Hanyang University / Microsoft Developer Technologies MVP / @CppKorea Founder / Reinforcement Learning KR Administrator
Kibua20 @kibua20
415 Followers 1K Following 현직 모바일 SW개발자가 실무 tips과 소식을 트윗해요~ 리트윗 환영합니다. 아래 블로그 방문해주세요. #우분투 #Python #안드로이드 #모바일광고 #REST #API
Ghost general @afeezawotundun
80 Followers 1K Following Am software engineer and fun lover , food lover
CodingADHD @CodingADHD
331 Followers 1K Following Backend / Distributed Systems Software Engineer with "severe" ADHD & Narcolepsy. Remote & WFH with my two lovely cats🏡
어엉부엉 @d_ijk_stra
15K Followers 3K Following 어엉부엉 삽니다. 실체는 '유'사'통'계업자지만 직함은 기계 학습 과학자. 제 트윗은 회사의 입장과는 무관합니다. 회사에선 기계를, 집에서는 아이를 키우느라 바쁩니다. 프로필 사진은 아이가 그려준 제 모습입니다.
asbubam @asbubam
5K Followers 3K Following Engineer. 독립일기: https://t.co/o0KziPHqpX Context-aware YouTube Summarizer: https://t.co/i5NTB98Phi
sen @sen
50K Followers 212 Following Earth from space in 4K 🛰️ . Democratizing space using video to inform, educate, inspire and benefit all humanity. 🌎 All videos © Sen Corporation Ltd.
Polymarket @Polymarket
1.8M Followers 6K Following The World's Largest Prediction Market. Trade politics, news, crypto, culture, sports, tech, & more. Discord: https://t.co/tzKrbDfF3x
파이토치 한국 �... @PyTorchKR
152 Followers 52 Following PyTorch Korea User Group, 한국어 커뮤니티의 새로운 글을 공유합니다. 📰 #PyTorchKR #파이토치 #파이토치한국사용자모임 --- 파이토치 한국 사용자 모임은 인공지능에 관심있는 분들께 PyTorch를 소개하고 함께 배우며 성장하는 것을 목표로 하고 있습니다🔥
Peter Kuhar @pkuhar
2K Followers 2K Following Entrepreneur, coder, maker. Mobile Health and Machine Learning. https://t.co/hjYo5bICZU
Dev Korea @dev_korea
155 Followers 2 Following Dev Korea is the #1 tech job platform and global tech community for English speakers in South Korea. Maker: @florianldt && @serin_heo
Open Design @OpenDesignHQ
12K Followers 57 Following The open-source Claude Design alternative. Local-first · BYOK · 64K+ GitHub stars. 22 CLIs · 150 systems · 100+ skills. Design for everyone.
ElevenLabs @ElevenLabs
186K Followers 11 Following AI research and products that transform how we interact with technology. Leading foundational models powering ElevenAgents, @ElevenCreative, and ElevenAPI.
Anthropic @AnthropicAI
1.5M Followers 2 Following We're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant @claudeai on https://t.co/FhDI3KQh0n.
stevibe @stevibe
27K Followers 1K Following LLM. Local AI addict. Building @BenchLocalAI Builds things nobody asked for. Benchmarks things for fun.
MiniMax (official) @MiniMax_AI
104K Followers 882 Following Code: @MiniMaxAgent Token Plan: https://t.co/BDCycxepZw API: https://t.co/fHRdSV7BwZ Community: https://t.co/uhxxfLgkLU
vLLM @vllm_project
44K Followers 36 Following A high-throughput and memory-efficient inference and serving engine for LLMs. Join https://t.co/lxJ0SfX5pJ to discuss together with the community!
Kimi.ai @Kimi_Moonshot
182K Followers 136 Following Built by Moonshot AI to empower everyone to be superhuman. @Kimidevs built for developers ⚡️API: https://t.co/XCrgjXAqMw DC: https://t.co/wBsBTn6Ncc
Runway @runwayml
283K Followers 331 Following Building AI to simulate the world. We're hiring: https://t.co/Aj11xygZYI
Midjourney @midjourney
460K Followers 0 Following A community supported research lab - exploring new mediums of thought and amplifying the imaginative powers of the human species.
Puneet Patwari @system_monarch
21K Followers 123 Following Principal @Atlassian | Helping engineers reach Staff/Principal | 1:1 Mentorship & Mock Interviews | 90+ System design fundamentals - https://t.co/Ots2nRhO5f
Maxime Rivest 🧙... @MaximeRivest
10K Followers 942 Following Searching for the wise way to interact with AI. Experiments are on GitHub Chat bot are likely not it. We probably need more AI commands and einks, less chats
Francesco Ciulla @FrancescoCiull4
211K Followers 499 Following - Head of DevRel Engineer at https://t.co/t5PMw0nLYM - Software Developer - Rust, Docker, Web Dev - Docker Captain - Keynote Speaker - Building a 1M community...66%
Dwarkesh Patel @dwarkesh_sp
242K Followers 1K Following Host of @dwarkeshpodcast https://t.co/3SXlu7fy6N https://t.co/4DPAxODFYi https://t.co/hQfIWdM1Un
A-Uta @UtaAoya
2K Followers 780 Following 個人PCでLocal LLMをどこまで動かせるか検証中。 RTX5070Ti / EVO-X2 / ミニPC / M5Stack / Stack-chan Minimal。 速度・設定・失敗例・再現条件を実測ベースで投稿します。
Donghee Na @dongheena92
2K Followers 375 Following - ML Infrastructure SWE @ Karrot (당근) - Python SC(2025-2026) and Core dev Opinions are my own not my employer https://t.co/hh9qPaDb7c
Ulanzi Japan(ウラ�... @UlanziJapan
10K Followers 261 Following カメラ、スマホなどの撮影アクセサリーメーカーUlanzi(ウランジ)の公式です! 新商品情報をお届け! フォローして最新情報とキャンペーンをチェック✨ TAG:#Ulanzi SNSおよびショップ:https://t.co/mi6Li0tpKG お問い合わせ:[email protected]
Everlier @Everlier
2K Followers 462 Following Building LLM agents & tools. Local inference will win. https://t.co/Z3gGqdXnXL Also: Facts / Mi / Skilled / Pace
Cosmic Tofu 🚐 Wish... @Team_CosmicTofu
1K Followers 774 Following In the Van, a cozy survival sim at the end of the world. Your van, your home, your road. 👇 Wishlist on Steam
Codex Releases @CodexReleases
19K Followers 3 Following Unofficial Codex release updates creds @OpenAI
Patrick Loeber @patloeber
73K Followers 1K Following member of technical staff @GoogleDeepMind • gemini api & ai studio • my views
AK @_akhaliq
511K Followers 3K Following AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo ,submit papers here: https://t.co/UzmYN5XOCi
Simon Willison @simonw
197K Followers 6K Following Creator @datasetteproj, co-creator Django. PSF board. Hangs out with @natbat. He/Him. Mastodon: https://t.co/t0MrmnJW0K Bsky: https://t.co/OnWIyhX4CH
Google Gemma @googlegemma
91K Followers 0 Following The official home of Google's Gemma. Lightweight, state-of-the-art open models by Google DeepMind, built on Gemini tech. What will you build? 🚀💻
Claude @claudeai
1.6M Followers 2 Following Claude is an AI assistant built by @anthropicai to be safe, accurate, and secure. Talk to Claude on https://t.co/ZhTwG8dz3D or download the app.






































