🏥 Quick Return to Emergency Room
You are in a specialist desk.
For full triage and doctors on duty, return here:
- WFGY Global Fix Map — main Emergency Room, 300+ structured fixes
- WFGY Problem Map 1.0 — 16 reproducible failure modes
Think of this page as a sub-room.
If you want full consultation and prescriptions, go back to the Emergency Room lobby.
This page is your hub to stabilize retrieval pipelines across popular vector stores.
If your results look similar but the answer is wrong, start here. Each store page gives guardrails, fix steps, and the same acceptance targets so you can verify without changing infra.
| Store | Best for | Why choose | Link |
|---|---|---|---|
| FAISS | local development, labs | fast, widely used, you manage it | faiss.md |
| Chroma | quick demos, notebooks | simple API, easy to start | chroma.md |
| Qdrant | production and multitenant | Rust core, good scaling, persistence | qdrant.md |
| Weaviate | hybrid search and schemas | first class filters, hybrid pipelines | weaviate.md |
| Milvus | enterprise ANN at scale | mature ecosystem and performance | milvus.md |
| pgvector | teams already on Postgres | keep data in the same DB, simple ops | pgvector.md |
| Redis (Search/Vec) | caches and small hybrid sets | key value plus vectors, low latency | redis.md |
| Elasticsearch (ANN) | text plus vector in one stack | reuse analyzers and infra you already have | elasticsearch.md |
| Pinecone | zero ops SaaS | managed reliability and steady API | pinecone.md |
| Typesense | simple full text plus vectors | friendly setup, good defaults | typesense.md |
| Vespa | large scale search and recsys | query routing and ranking at scale | vespa.md |
- High similarity but wrong meaning.
- Citations do not match the retrieved section.
- Hybrid retrieval performs worse than a single retriever.
- After deploy, query casing or analyzer or metric does not line up.
- Index looks healthy but coverage stays low.
- ΔS(question, retrieved) ≤ 0.45
- Coverage of target section ≥ 0.70
- λ_observe convergent across three paraphrases
- E_resonance flat on long windows
-
Embedding ≠ Semantic
Wrong meaning despite high similarity.
→ embedding-vs-semantic.md -
Retrieval traceability
Snippet or section mismatch, unverifiable citations.
→ retrieval-traceability.md
Payload contract → data-contracts.md -
Ordering or version skew
Runtime loads the wrong index or analyzer.
→ bootstrap-ordering.md · predeploy-collapse.md -
Hybrid collapse or query split
HyDE and BM25 disagree, reranker blind spots.
→ Pattern → pattern_query_parsing_split.md
→ Knobs → rerankers.md
-
Lock metrics and analyzers
One embedding model per field. One distance function. Same analyzer for write and read. -
Contract the snippet
Require{snippet_id, section_id, source_url, offsets, tokens}and enforce cite then explain.
→ data-contracts.md -
Add deterministic reranking
Keep candidate lists from BM25 and ANN. Detect query split.
→ rerankers.md -
Cold start and deploy fences
Block traffic until index hash, analyzer, and model versions match.
→ bootstrap-ordering.md -
Observability
Log ΔS and λ across retrieve, rerank, reason. Alert when ΔS ≥ 0.60. -
Regression gate
Require coverage ≥ 0.70 and ΔS ≤ 0.45 before publish.
I uploaded TXT OS and the WFGY Problem Map pages.
Store: <name>. Retrieval: <bm25|ann|hybrid> with <distance>.
Audit this query and return:
- ΔS(question,retrieved) and λ across retrieve → rerank → reason.
- If ΔS ≥ 0.60, choose one minimal structural fix and name the page:
embedding-vs-semantic, retrieval-traceability, data-contracts, rerankers.
- JSON only:
{ "citations":[...], "ΔS":0.xx, "λ":"→|←|<>|×", "next_fix":"..." }| Tool | Link | 3 step setup |
|---|---|---|
| WFGY 1.0 PDF | Engine Paper | 1) Download 2) Upload to your LLM 3) Ask “Answer using WFGY + ” |
| TXT OS (plain text OS) | TXTOS.txt | 1) Download 2) Paste into any LLM chat 3) Type “hello world” to boot |
| Layer | Page | What it’s for |
|---|---|---|
| ⭐ Proof | WFGY Recognition Map | External citations, integrations, and ecosystem proof |
| ⚙️ Engine | WFGY 1.0 | Original PDF tension engine and early logic sketch (legacy reference) |
| ⚙️ Engine | WFGY 2.0 | Production tension kernel for RAG and agent systems |
| ⚙️ Engine | WFGY 3.0 | TXT based Singularity tension engine (131 S class set) |
| 🗺️ Map | Problem Map 1.0 | Flagship 16 problem RAG failure taxonomy and fix map |
| 🗺️ Map | Problem Map 2.0 | Global Debug Card for RAG and agent pipeline diagnosis |
| 🗺️ Map | Problem Map 3.0 | Global AI troubleshooting atlas and failure pattern map |
| 🧰 App | TXT OS | .txt semantic OS with fast bootstrap |
| 🧰 App | Blah Blah Blah | Abstract and paradox Q&A built on TXT OS |
| 🧰 App | Blur Blur Blur | Text to image generation with semantic control |
| 🏡 Onboarding | Starter Village | Guided entry point for new users |
If this repository helped, starring it improves discovery so more builders can find the docs and tools.
