Aevum Encyclopedia / OpenFaaS

Total Invocations (24h)

14.2M

↑ 8.4% vs previous day

Avg Latency (p95)

142ms

↓ 12ms optimized cold starts

Success Rate

99.94%

Auto-retry enabled on 429/503

Active Functions

18 deployed, 6 scaling

📦 Registered Functions

/fact-verify

deployed

Cross-references claims against primary sources using LLM-augmented RAG pipeline. Returns confidence score & citation links.

Latency: 89ms Scale: 1-12 Inv/min: 4,200

/semantic-search

scaling

Dense vector retrieval over 2.4M indexed articles. Supports multi-lingual query embedding and reranking.

Latency: 210ms Scale: 3-24 Inv/min: 8,100

/cross-translate

deployed

Neural MT with domain-specific fine-tuning for academic, historical, and technical terminology preservation.

Latency: 145ms Scale: 1-8 Inv/min: 2,400

/graph-indexer

deployed

Extracts entities & relationships, updates Neo4j knowledge graph, and triggers dependency resolution.

Latency: 320ms Scale: 1-6 Inv/min: 950

/content-moderate

scaling

Pre-commit validation for contributor submissions. Checks for plagiarism, policy violations, and structural compliance.

Latency: 110ms Scale: 2-10 Inv/min: 1,800

/citation-resolver

deployed

Resolves DOIs, arXiv IDs, and ISBNs. Fetches metadata, verifies publisher stamps, and formats output in Chicago/APA/MLA.

Latency: 75ms Scale: 1-4 Inv/min: 3,100

📡 API Reference & Invocation

                    
curl -X POST https://gateway.aevum.dev/openfaas/function/fact-verify \
  -H "Authorization: Bearer $AE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "claim": "Quantum entanglement violates local realism.",
    "language": "en",
    "depth": "high"
  }'
                

                    
const res = await fetch("https://gateway.aevum.dev/openfaas/function/semantic-search", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.AE_API_KEY}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    query: "Renaissance art techniques",
    limit: 10,
    rerank: true
  })
});
const results = await res.json();
        

                    
import requests

resp = requests.post(
    "https://gateway.aevum.dev/openfaas/function/cross-translate",
    headers={"Authorization": f"Bearer {AE_API_KEY}"},
    json={
        "text": "The knowledge graph has been updated.",
        "src_lang": "en",
        "tgt_lang": "ja"
    }
)
print(resp.json()["translated"])
        

                    
# Authentication requires a valid Bearer token issued by the Aevum IAM service.
# Tokens are scoped to function namespaces and rate-limited per key.

Header: Authorization: Bearer <api_key>
Scopes: openfaas:invoke, openfaas:read, openfaas:logs
Rate Limit: 500 req/min (standard), 5000 req/min (enterprise)
MFA: Enforced for production namespace access
        

📜 Live Deployment Log

14:02:11 [DEPLOY] semantic-search: v2.4.1 rolled out to eu-west-1

14:01:45 [SCALE] fact-verify: replicas 4 → 7 (trigger: queue_depth > 80)

14:00:33 [WARN] cross-translate: cold start detected (320ms), warming pool engaged

13:58:12 [HEALTH] gateway: 24/24 functions responding (p99: 142ms)

13:55:01 [DEPLOY] citation-resolver: v1.8.0 checksum verified, traffic shifted 100%

13:52:44 [RATE] content-moderate: key_ae_9x2m hit 429 limit, throttled for 60s

13:50:19 [SYNC] knowledge-graph: 14,202 edges committed to Neo4j cluster