RL//SEARCHENTERPRISE KNOWLEDGE COMMAND
S100--:--:--OVERVIEW
LIVE
ELASTICSEARCH • MEILISEARCH • FULL-TEXT • SEMANTIC SEARCH • AI SEARCH • HYBRID RETRIEVAL • RANKING • FILTERS • FACETS • ANALYTICS • INDEXING • PROJECTS • DOCUMENTATION • ARTICLES • REPOSITORIES •
ROBERTO UNIFIED ENTERPRISE SEARCH CONTROL PLANE
ELASTICSEARCH / MEILISEARCH / HYBRID RETRIEVAL

ONESEARCH PLANE.EVERY KNOWLEDGE SOURCE.

Search every project, document, article, and repository through one responsive system that combines enterprise retrieval, instant search, semantic meaning, ranking policy, filters, indexing, and analytics.

—Indexed docs
2Search engines
4Search modes
—Last latency
01

Unified search architecture

One query contract routes work to the best engine and merges results into a normalized response.

SOURCESProjects · Docs · Articles · Repositories
→
SEARCH ORCHESTRATORQuery analysis · routing · fusion · policy
→
→
RESULT EXPERIENCERank · filter · highlight · answer
02

Live enterprise search

Try lexical, semantic, hybrid, or AI search across the portfolio knowledge index.

SEARCH QUERYLIVE API
SEARCH RESPONSEREADY
Run a search to generate a grounded summary.
The ranked result stream will appear here.
03

Search capabilities

Every capability is visible, testable, and connected to the same index contract.

FULL-TEXT

Analyzers, field boosts, highlights, typo handling, phrase and prefix matching.

SEMANTIC

Meaning-aware retrieval using vector representations and hybrid score fusion.

AI SEARCH

Grounded answers synthesized from top-ranked, attributable result passages.

ANALYTICS

Query volume, zero-result searches, latency, clicks, and content coverage.

S200

Documentation

Unified Search API, deployment topology, indexing lifecycle, and production integration guides.

API CONTRACTOPENAPI 3.1
POST /api/search
{
  "query": "AI architecture",
  "mode": "hybrid",
  "filters": { "type": "project" },
  "engine": "auto"
}
Open specification ↗
DEPLOYMENTDOCKER
docker compose up -d

Elasticsearch :9200
Meilisearch   :7700
Website       : Cloudflare Pages
Open compose file ↗
SEARCH ENDPOINTS10 ROUTES
MethodPathPurpose
GET/api/healthRuntime health
GET/api/search/overviewIndex topology
POST/api/searchUnified search
GET/api/searchQuery-string search
POST/api/index/documentsIndex a document
GET/api/index/statusIndex status
GET/api/analyticsSearch analytics
POST/api/analytics/eventsTrack interaction
GET/api/github/repositoriesOfficial repos
GET/api/facetsAvailable filters
S300

GitHub

Official source repositories powering the enterprise and instant search layers.

elastic/elasticsearch

Distributed search and analytics engine with full-text, vector, and production-scale retrieval.

Open GitHub ↗

meilisearch/meilisearch

Fast search engine API with typo tolerance, facets, semantic and hybrid search.

Open GitHub ↗
S310

Elasticsearch

The enterprise retrieval, analytics, and vector-search backbone for large-scale portfolio knowledge.

ES
DISTRIBUTED SEARCH + ANALYTICS + VECTOR DATABASE

Elasticsearch

Handles rich Query DSL, analyzers, aggregations, vector retrieval, hybrid fusion, near-real-time indexing, and search analytics for production-scale workloads.

Official documentation ↗
QUERY DSL

Match, multi-match, phrase, boolean, prefix, fuzzy, and structured queries.

SEMANTIC TEXT

Vector and inference workflows for semantic and hybrid retrieval.

AGGREGATIONS

Facets, distributions, trends, and analytics over indexed content.

SCALABILITY

Distributed indices, shards, replicas, bulk ingestion, and near-real-time search.

S320

Meilisearch

The fast, typo-tolerant search experience for responsive website search and instant discovery.

M
INSTANT FULL-TEXT + SEMANTIC + HYBRID SEARCH

Meilisearch

Serves responsive search-as-you-type, typo tolerance, ranking rules, filterable attributes, facets, and fast result delivery for user-facing discovery.

Official documentation ↗
INSTANT SEARCH

Low-latency results designed for interactive website search.

TYPO TOLERANCE

Relevant results even when the query contains common spelling mistakes.

RANKING RULES

Words, typo, proximity, attribute rank, sort, word position, and exactness.

FILTERS + FACETS

Filterable attributes enable fast, structured exploration of the index.

S400

Full-text Search

Lexical retrieval with analyzers, boosts, highlighting, phrase matching, and typo-aware instant results.

LEXICAL SEARCH LABLIVE
RANKED HITSREADY
Search results will appear here.
S500

Semantic Search

Retrieve concepts by meaning, not only exact terms, then blend vector similarity with lexical confidence.

SEMANTIC QUERYVECTOR MODE
HYBRID SCORE0.58 × semantic + 0.32 × lexical + 0.10 × quality
SEMANTIC MATCHESREADY
Meaning-aware matches will appear here.
S600

AI Search

Grounded answer generation over retrieved portfolio evidence with traceable sources.

ASK THE KNOWLEDGE INDEXRAG DEMO
GROUNDING POLICY

The demo answer is generated only from retrieved index records and always returns supporting result cards.

ANSWER + EVIDENCEREADY
Ask a question to generate a grounded response.
S700

Ranking

Control relevance using lexical confidence, semantic similarity, freshness, popularity, exactness, and content quality.

RANKING POLICYINTERACTIVE
EXPLAIN SCORETOP RESULTS
Run a test to inspect score components.
FIELD BOOSTS

Title, tags, technology, description, and body contribute different lexical weights.

RECIPROCAL FUSION

Merge lexical and vector result lists without requiring identical raw scores.

BUSINESS SIGNALS

Freshness, popularity, featured status, and content completeness can break ties.

EXPLAINABILITY

Each result exposes lexical, semantic, and quality score components.

S800

Filters

Facets and constraints narrow the unified index without sacrificing relevance or responsiveness.

FILTER BUILDERFACETED
FILTERED RESULTSREADY
S900

Analytics

Measure query demand, latency, zero-result searches, mode adoption, and content coverage.

0Total searches
0 msAverage latency
0%Zero-result rate
0Indexed documents
SEARCH MODE DISTRIBUTIONLIVE
RECENT QUERIES0 EVENTS
Run searches to populate analytics.
S910

Indexing

Normalize and ingest projects, documentation, articles, and repositories into both search engines.

INDEX DOCUMENTLIVE API
INDEX STATUSLOADING
Submit a document to inspect the normalized indexing response.
01 / EXTRACTGitHub · Markdown · CMS · API
→
02 / NORMALIZESchema · metadata · ACL
→
03 / ENRICHChunks · tags · embeddings
→
04 / INDEXElasticsearch + Meilisearch
→
05 / VERIFYCounts · freshness · quality