AICommunity
Log in Sign up 한국어

#rag

Hot New Top
Popular tags #rag#RAG#opensource#agents#llm#에이전트#임베딩#비용#GPU#오픈소스#커뮤니티#serving
  1. 9
    votes
    A We replaced our support team's KB with a RAG bot
    Ticket deflection went from 0% to 38% in a month. Here's the architecture and the mistakes we made.
    Showcase @alexus · 1일 전 · comments 1 #support#rag#case-study
  2. 5
    votes
    P RAG eval: how do you measure answer quality?
    We use RAGAS but the scores don't correlate with human judgment. What do you actually use in production?
    Q&A @polyglot · 1일 전 · comments 1 #rag#evaluation#ragas
  3. 6
    votes
    K Best vector DB for a solo dev in 2026?
    Chroma vs Qdrant vs LanceDB for a small self-hosted RAG service. Disk and RAM matter — running on a free-tier VM.
    Q&A @kai · 1일 전 · comments 2 #vectordb#rag
  4. 8
    votes
    V My 8B model beats our old fine-tune with RAG only
    Replaced a custom fine-tune with a generic 8B + good retrieval. Cheaper to maintain, better scores. Data point for you.
    Showcase @vectorlee · 1일 전 · comments 1 #rag#finetune#case-study
  5. 6
    votes
    R How do you handle hallucinated citations in RAG?
    Model cites chunks that don't actually support the answer. Any post-processing tricks that work?
    Q&A @rai · 1일 전 · comments 0 #rag#hallucination#citations
  6. 8
    votes
    J My RAG chatbot for a 10k-page policy manual actually works
    Our org has 10k pages of policy PDFs nobody reads. Built a RAG bot with citations that links back to the exact page. Support queries dropped noticeably and compliance is happy because everything is traceable. Side project that became the team's favorite tool.
    Showcase @Jennifer Chen · 2일 전 · comments 0 #rag#chatbot#showcase
  7. 3
    votes
    S RAG vs fine-tuning for a 5k-doc internal wiki — which actually worked for you?
    We need to ground answers in internal docs that update weekly. RAG feels obvious but retrieval quality is killing us. Fine-tuning is a one-time cost but the docs change. Anyone run a hybrid? What's your split?
    Q&A @Sofia Alvarez · 1일 전 · comments 0 #rag#finetuning#grounding
  8. 6
    votes
    A Built a tiny eval harness for our RAG pipeline — 100 lines, huge ROI
    Link
    We were merging PRs blind. Now every RAG change runs 40 questions with expected answers and fails the build if recall drops. It's crude but it caught two regressions last week. Sometimes the simple stuff wins.
    Tools/Projects @Anna Nowak · 2일 전 · comments 0 #rag#evals#testing
  9. 4
    votes
    D How do you handle 200k+ context without losing your mind (and your money)?
    Long context is expensive and my tests show the model forgets the middle anyway. Chunking into a RAG loop works but feels like giving up. Do you just pay for big context or actually engineer around it?
    Q&A @David Kim · 2일 전 · comments 0 #context#rag#llm
  10. 6
    votes
    D Open-sourced my 'second brain' — 200k notes in a vector DB
    Link
    Been dumping notes, highlights, and bookmarks into a local RAG setup for two years. Search is instant and surprisingly good. Open-sourced the whole thing so it's reproducible. Warning: your old notes are more embarrassing than you remember.
    Showcase @David Kim · 6일 전 · comments 0 #second-brain#rag#opensource
  11. 3
    votes
    T Interviewer asked me to design a RAG system on a whiteboard. I froze.
    In hindsight it was a fair question but I panicked and started rambling about chunk sizes. I've built two RAG systems in production. Doesn't matter, whiteboard me is a different person. Anyone else have interview horror stories?
    Free @Tom Hall · 6일 전 · comments 0 #interview#rag#humor
  12. 5
    votes
    D Multi-tenant RAG: one big index or per-tenant collections?
    We're about to onboard our first enterprise customers and I have to decide. Shared index with metadata filters seems simpler but I'm scared of cross-tenant leakage. Per-tenant collections mean more ops. What do you run?
    Q&A @David Kim · 10일 전 · comments 0 #rag#multi-tenant#security
  13. 12
    votes
    H 사내 문서 Q&A 챗봇, 3주 만에 배포했습니다
    Image Link
    사내 매뉴얼 2천 페이지를 RAG로 묶어서 챗봇을 만들었습니다. 임베딩은 bge-m3, 서빙은 vLLM으로 했고, 정확도는 사내 QA 기준 87% 나옵니다.
    Showcase @hanbit · 7시간 전 · comments 1 #RAG#사내#배포
  14. 8
    votes
    R RAG 청크 크기, 문서 유형별로 어떻게 다르게 두시나요?
    Image Link
    코드 문서는 300토큰, 일반 문서는 500~800토큰으로 쓰고 있는데요. 법률 문서처럼 구조화된 건 섹션 단위로 자르는 게 낫다는 의견도 있더라고요. 다들 어떻게 하세요?
    Q&A @rai · 7시간 전 · comments 1 #RAG#청크#임베딩
  15. 9
    votes
    S 폐쇄망 RAG 키트 — 실제 배포 사례 공유
    Image Link
    인터넷이 안 되는 환경에서 LLM+RAG를 돌려야 하는 프로젝트였는데, Ollama + 로컬 임베딩으로 구성했습니다. 오프라인 배포에 관심 있으신 분들께 도움이 되길.
    Showcase @sean · 9시간 전 · comments 1 #오프라인#RAG#사례
  16. 13
    votes
    P 파인튜닝 vs RAG — 언제 뭘 써야 하나요?
    Image Link
    도메인 지식이 필요하면 RAG, 스타일/포맷 고정이면 FT라는 기본 공식이 있는데, 실제 프로젝트에서는 어떤 기준으로 결정하시나요?
    Q&A @polyglot · 14시간 전 · comments 0 #RAG#파인튜닝#아키텍처
  17. 10
    votes
    S Docker + Ollama + FastAPI RAG 템플릿 공유합니다
    Image Link
    로컬 RAG 서비스 최소 구성 템플릿을 만들어봤습니다. docker compose 하나로 Ollama + FastAPI + ChromaDB가 뜹니다. 피드백 환영해요.
    Tools/Projects @sean · 12시간 전 · comments 1 #RAG#템플릿#Docker
  18. 11
    votes
    A 우리 팀 KB를 RAG 봇으로 대체했습니다
    Image Link
    지식베이스 검색 시간이 하루 2시간 → 10분으로 줄었습니다. 초기엔 환각이 문제였는데, 인용 소스 표시를 강제하니 신뢰도가 확 올라갔어요.
    Showcase @alexus · 18시간 전 · comments 1 #KB#RAG#생산성
  19. 15
    votes
    S airgap RAG 키트 만들었습니다
    폐쇄망용 완전 오프라인 RAG. Ollama만으로 동작, 임베딩은 bge-m3. 피드백 환영합니다.
    Tools/Projects @sean · 1일 전 · comments 3 #RAG#폐쇄망#Ollama
  20. 8
    votes
    S 폐쇄망 RAG 키트 — 실제 배포 사례 공유
    금융권 고객사에 설치했습니다. 인터넷 완전 차단 환경에서도 임베딩+검색+생성 전부 동작. 자세한 내용은 글에.
    Showcase @sean · 1일 전 · comments 1 #폐쇄망#RAG#배포
  21. 8
    votes
    V What embedding model do you use for RAG?
    I'm working on Korean document retrieval and deciding between bge-m3 and multilingual-e5. Please share your experiences.
    Q&A @vectorlee · 1일 전 · comments 4 #RAG#임베딩
  22. 8
    votes
    G 파인튜닝 vs RAG, 언제 뭘 써야 하나요?
    도메인 용어가 많은데 파인튜닝이 나을지 RAG가 나을지. 둘 다 해보신 분들 의견 궁금합니다.
    Q&A @gptkim · 1일 전 · comments 1 #파인튜닝#RAG#선택
  23. 7
    votes
    M 임베딩 모델 교체하면 기존 데이터 다시 임베딩해야 하나요?
    bge-m3에서 다른 모델로 바꾸려는데, 기존 문서 전부 다시 인덱싱해야 하는 게 맞는지요?
    Q&A @mlpark · 1일 전 · comments 1 #임베딩#인덱싱#RAG