[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79SADA55ZCRGEJJS4PJ2GPY":3,"skill-same:01M79SADA55ZCRGEJJS4PJ2GPY":53},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":32,"commitSha":33,"skillDirectory":34,"forksCount":14,"repositoryHtmlUrl":35,"repositoryHomepageUrl":6,"ownerHtmlUrl":36,"ownerBlogUrl":6,"ownerDescription":37,"updatedAt":38,"install":39,"resources":52},"01M79SADA55ZCRGEJJS4PJ2GPY","vector-store-architect",null,"Architect-level vector database design — index type and recall\u002Flatency trade-offs, filtering and tenancy, metadata schema, sharding and replication, updates and deletes, and the operational difference between pgvector, Qdrant, Weaviate, and managed services. Use when vectors need a home beyond a prototype.","SriAkhilSJ\u002FPulseAIRepo","SriAkhilSJ","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F206201543?v=4","01M79RWFD1Q09B89SXTTKKJY6Z","Sri Akhil SJ","01M79RWFE26NPG0BZ1FRNX376H",0,"MIT","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fai-ml-architect\u002Fretrieval-architect\u002Fvector-store-architect\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: vector-store-architect\ndescription: Architect-level vector database design — index type and recall\u002Flatency trade-offs, filtering and tenancy, metadata schema, sharding and replication, updates and deletes, and the operational difference between pgvector, Qdrant, Weaviate, and managed services. Use when vectors need a home beyond a prototype.\nlicense: MIT\ncompatibility: pgvector, Qdrant, Weaviate, Milvus, Pinecone, OpenSearch\nmetadata:\n  crusor-domain: ai-ml-architect\n  crusor-path: ai-ml-architect\u002Fretrieval-architect\u002Fvector-store-architect\n  crusor-origin: original\n  crusor-sources: \"github.com\u002Fpgvector\u002Fpgvector; qdrant.tech\u002Fdocumentation; weaviate.io\u002Fdevelopers\u002Fweaviate; docs.pinecone.io\"\n  crusor-triggers: \"vector database, pgvector, qdrant, weaviate, pinecone, hnsw, ivf, ann index, similarity search, vector index, vector search, irrelevant results, retrieval quality\"\n  crusor-verified-on: \"2026-10-30\"\n---\n\n# Vector Store Architect\n\nA vector store is **an index with a consistency model**. Most failures are not\nrecall problems — they are filtering, tenancy, and delete semantics.\n\n## When to use\n\n- More than a few hundred thousand vectors, or strict latency targets.\n- Metadata filtering combined with similarity search.\n- Multi-tenant retrieval with isolation requirements.\n\n## When NOT to use\n\n- Small corpus where Postgres with pgvector is already deployed and sufficient.\n- Pure keyword search — that is a text index.\n\n## Decision 1 — Engine\n\n| Engine | Fit |\n|---|---|\n| pgvector | already on Postgres, moderate scale, transactional consistency |\n| Qdrant | filtering-heavy, self-hosted, strong payload model |\n| Weaviate | hybrid search and modules out of the box |\n| Milvus | very large scale, dedicated ops |\n| Pinecone | managed, low ops, higher cost at scale |\n| OpenSearch\u002FElasticsearch | existing search platform, hybrid needs |\n\n- Introducing a new datastore means a new backup, auth, and monitoring story. Pay\n  it only for a real capability gap.\n- Estimate: vectors × dimensions × 4 bytes × (1 + index overhead) — then compare to\n  available RAM.\n\n## Decision 2 — Index\n\n| Index | Trade-off |\n|---|---|\n| flat\u002Fexact | perfect recall, slow at scale |\n| HNSW | fast queries, memory-hungry, high build cost |\n| IVF-PQ | memory efficient, recall loss, tuning required |\n| disk-based (DiskANN) | large corpora with acceptable latency |\n\n- Always measure recall@k against exact search on a sample; \"it feels fast\" is not\n  recall.\n- Tune `m`\u002F`ef_construction`\u002F`nprobe` against latency budget, then freeze the\n  parameters with the index version.\n- Rebuilding an index is an operational event: plan capacity and a maintenance\n  window.\n\n## Decision 3 — Schema and filtering\n\n- Metadata carries: tenant, document id, section, timestamps, permissions,\n  embedding model version.\n- Filter **inside** the vector search (pre-filter), not after retrieval — post-filter\n  silently returns fewer than k results.\n- Pre-filtering with high-selectivity filters can be slow on HNSW; check the\n  engine's guidance.\n- Store the text or a pointer to it; you will need to re-embed and to show snippets.\n\n## Decision 4 — Tenancy\n\n| Model | Isolation |\n|---|---|\n| one collection, tenant in payload | weakest, filter must be enforced everywhere |\n| collection per tenant | stronger, many collections to manage |\n| namespace\u002Fpartition per tenant | balance, engine-dependent |\n| database\u002Fcluster per tenant | strongest, most expensive |\n\n- Enforce the tenant filter in the data-access layer, never by trusting the caller's\n  query.\n- Test cross-tenant leakage as a first-class test case.\n\n## Decision 5 — Writes and deletes\n\n- Upsert by content hash + document id; deletion by document id must remove all its\n  chunks.\n- Deletes are often deferred\u002Fasync — verify the vector is actually gone before\n  claiming deletion.\n- Versioning: keep the embedding model id on each vector to make migration and\n  audit possible.\n- Snapshot\u002Fbackup strategy: a rebuilt index from source text is preferable to\n  backing up vectors, if ingestion is fast.\n\n## Failure modes at scale\n\n- Post-filter dropping results: \"top 5\" often returns 2.\n- One collection, tenant filter forgotten in a new code path; data leaks.\n- HNSW index in RAM exceeding the node size; OOM under load.\n- Deleted documents still returned weeks later because deletes were queued.\n- Recall degraded after an index rebuild with default parameters.\n\n## Verification receipts\n\n1. Engine choice with the capability that justified a new datastore.\n2. Recall@k measured against exact search on a sample.\n3. Filter selectivity test: filtered queries still return k results.\n4. Cross-tenant test proving isolation.\n5. Delete verification: document removed from every chunk and result.\n","# Vector Store Architect\n\nA vector store is **an index with a consistency model**. Most failures are not\nrecall problems — they are filtering, tenancy, and delete semantics.\n\n## When to use\n\n- More than a few hundred thousand vectors, or strict latency targets.\n- Metadata filtering combined with similarity search.\n- Multi-tenant retrieval with isolation requirements.\n\n## When NOT to use\n\n- Small corpus where Postgres with pgvector is already deployed and sufficient.\n- Pure keyword search — that is a text index.\n\n## Decision 1 — Engine\n\n| Engine | Fit |\n|---|---|\n| pgvector | already on Postgres, moderate scale, transactional consistency |\n| Qdrant | filtering-heavy, self-hosted, strong payload model |\n| Weaviate | hybrid search and modules out of the box |\n| Milvus | very large scale, dedicated ops |\n| Pinecone | managed, low ops, higher cost at scale |\n| OpenSearch\u002FElasticsearch | existing search platform, hybrid needs |\n\n- Introducing a new datastore means a new backup, auth, and monitoring story. Pay\n  it only for a real capability gap.\n- Estimate: vectors × dimensions × 4 bytes × (1 + index overhead) — then compare to\n  available RAM.\n\n## Decision 2 — Index\n\n| Index | Trade-off |\n|---|---|\n| flat\u002Fexact | perfect recall, slow at scale |\n| HNSW | fast queries, memory-hungry, high build cost |\n| IVF-PQ | memory efficient, recall loss, tuning required |\n| disk-based (DiskANN) | large corpora with acceptable latency |\n\n- Always measure recall@k against exact search on a sample; \"it feels fast\" is not\n  recall.\n- Tune `m`\u002F`ef_construction`\u002F`nprobe` against latency budget, then freeze the\n  parameters with the index version.\n- Rebuilding an index is an operational event: plan capacity and a maintenance\n  window.\n\n## Decision 3 — Schema and filtering\n\n- Metadata carries: tenant, document id, section, timestamps, permissions,\n  embedding model version.\n- Filter **inside** the vector search (pre-filter), not after retrieval — post-filter\n  silently returns fewer than k results.\n- Pre-filtering with high-selectivity filters can be slow on HNSW; check the\n  engine's guidance.\n- Store the text or a pointer to it; you will need to re-embed and to show snippets.\n\n## Decision 4 — Tenancy\n\n| Model | Isolation |\n|---|---|\n| one collection, tenant in payload | weakest, filter must be enforced everywhere |\n| collection per tenant | stronger, many collections to manage |\n| namespace\u002Fpartition per tenant | balance, engine-dependent |\n| database\u002Fcluster per tenant | strongest, most expensive |\n\n- Enforce the tenant filter in the data-access layer, never by trusting the caller's\n  query.\n- Test cross-tenant leakage as a first-class test case.\n\n## Decision 5 — Writes and deletes\n\n- Upsert by content hash + document id; deletion by document id must remove all its\n  chunks.\n- Deletes are often deferred\u002Fasync — verify the vector is actually gone before\n  claiming deletion.\n- Versioning: keep the embedding model id on each vector to make migration and\n  audit possible.\n- Snapshot\u002Fbackup strategy: a rebuilt index from source text is preferable to\n  backing up vectors, if ingestion is fast.\n\n## Failure modes at scale\n\n- Post-filter dropping results: \"top 5\" often returns 2.\n- One collection, tenant filter forgotten in a new code path; data leaks.\n- HNSW index in RAM exceeding the node size; OOM under load.\n- Deleted documents still returned weeks later because deletes were queued.\n- Recall degraded after an index rebuild with default parameters.\n\n## Verification receipts\n\n1. Engine choice with the capability that justified a new datastore.\n2. Recall@k measured against exact search on a sample.\n3. Filter selectivity test: filtered queries still return k results.\n4. Cross-tenant test proving isolation.\n5. Delete verification: document removed from every chunk and result.\n",{"name":5,"license":15,"metadata":24,"description":7,"compatibility":31},{"crusor-path":25,"crusor-domain":26,"crusor-origin":27,"crusor-sources":28,"crusor-triggers":29,"crusor-verified-on":30},"ai-ml-architect\u002Fretrieval-architect\u002Fvector-store-architect","ai-ml-architect","original","github.com\u002Fpgvector\u002Fpgvector; qdrant.tech\u002Fdocumentation; weaviate.io\u002Fdevelopers\u002Fweaviate; docs.pinecone.io","vector database, pgvector, qdrant, weaviate, pinecone, hnsw, ivf, ann index, similarity search, vector index, vector search, irrelevant results, retrieval quality","2026-10-30","pgvector, Qdrant, Weaviate, Milvus, Pinecone, OpenSearch","main","8fa4e240be65b98472c9b2d3a505ade2aea1a5c6","skills\u002Fai-ml-architect\u002Fretrieval-architect\u002Fvector-store-architect","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ","Studying in CMR institute of Technology\r\n2024 - 2028","2026-11-13 21:02:23",{"source":8,"skillName":5,"commands":40,"prompt":50,"zipAvailable":51},[41,44,47],{"tool":42,"command":43},"npx","npx skills add SriAkhilSJ\u002FPulseAIRepo --skill vector-store-architect",{"tool":45,"command":46},"bunx","bunx skills add SriAkhilSJ\u002FPulseAIRepo --skill vector-store-architect",{"tool":48,"command":49},"pnpm","pnpm dlx skills add SriAkhilSJ\u002FPulseAIRepo --skill vector-store-architect","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fai-ml-architect\u002Fretrieval-architect\u002Fvector-store-architect\u002FSKILL.md",true,[],[54,61,68,75,82,89],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M79TTAJNACCNXCEN3649CMMD","interaction-patterns-architect","Architect-level interaction design in code — choosing established patterns over invented ones, feedback and latency behaviour, destructive-action safeguards, navigation and overlay conventions, and the state model behind every interactive surface. Use when a screen has to behave, not just look.","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fwebsite-architect\u002Fui-ux-implementation-architect\u002Finteraction-patterns-architect\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M79TT2ESKB1TKZBYY9KYJHV9","design-system-ops-architect","Architect-level operation of a design system in code — token pipelines and versioning, component API and variant policy, the catalogue as the contract, adoption metrics, and the governance that stops a system from forking into three. Use when a component library must survive several teams.","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fwebsite-architect\u002Fui-ux-implementation-architect\u002Fdesign-system-ops-architect\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M79TSV0PWGNK3PRX9P4JDR9N","design-handoff-architect","Architect-level design handoff — what must be specified before implementation starts, tokens over raw values, responsive and content-limit contracts, state matrices, and the review loop that catches drift between design and code. Use when a design is about to become code.","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fwebsite-architect\u002Fui-ux-implementation-architect\u002Fdesign-handoff-architect\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M79TSJS9BFXNP696FMJ9GKDA","content-microcopy-architect","Architect-level UI content — where strings live, voice and terminology consistency, error and empty-state copy as part of the interface, translation-ready authoring, and the content model a product needs before it multiplies screens. Use when the words in the interface matter as much as the pixels.","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fwebsite-architect\u002Fui-ux-implementation-architect\u002Fcontent-microcopy-architect\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M79TSCNBST3N1M68V1KPH9HQ","ui-ux-implementation-architect","Architect-level direction for turning a design into working UI — the handoff contract, component ownership versus screen-level layout, interaction and state conventions, design QA gates, and the review habits that keep an implementation faithful without frozen pixels. Use before any specific UI implementation skill.","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fwebsite-architect\u002Fui-ux-implementation-architect\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M79TS4N6A4D2QQ433VXS5D0R","test-strategy-architect","Architect-level test strategy — allocating assertions across unit, integration, contract and end-to-end layers, designing for testability through dependency boundaries, fixture and data strategy, and the flake and coverage policies that keep a suite trustworthy. Use when defining how a codebase is tested.","https:\u002F\u002Fgithub.com\u002FSriAkhilSJ\u002FPulseAIRepo\u002Fblob\u002Fmain\u002Fskills\u002Fwebsite-architect\u002Ftesting-and-deployment-architect\u002Ftest-strategy-architect\u002FSKILL.md",[],[]]