Publish artifacts for production-ai-observability-20260820

README.md

@@ -0,0 +1,59 @@
+---
+license: cc-by-4.0
+language:
+- en
+pretty_name: Production AI Observability Monitor Synthetic Evaluation Set
+size_categories:
+- n<1K
+task_categories:
+- text-classification
+tags:
+- synthetic
+- ai-observability
+- evaluation
+- text-classification
+- token-classification
+- summarization
+- zero-shot-classification
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train.jsonl
+  - split: test
+    path: data/test.jsonl
+---
+
+# Production AI Observability Monitor Synthetic Dataset
+
+## Summary
+
+This dataset contains 14 training examples and 4
+held-out examples for **Production AI teams need trace-level signals for latency, token growth, tool failures, and low-quality outputs.**
+
+Every record is synthetic and includes:
+
+- `input`: query, event, or feature description
+- `label`: expected class, route, relation, or evidence category
+- `context`: synthetic supporting context
+- `source`: fictional source identifier
+- `variant`: generation pattern
+- `synthetic`: always `true`
+
+## Uses
+
+- Reproducible unit and integration tests
+- Baseline model training
+- Evaluation harness development
+- Schema and architecture demonstrations
+
+## Limitations
+
+Thresholds are demonstration defaults and need calibration against each production workload.
+
+This dataset does not represent real users, patients, customers, production
+traffic, or licensed media. It must not be presented as real-world evidence.
+
+## Related Model
+
+[RKB109/production-ai-observability-20260820-model](https://huggingface.co/RKB109/production-ai-observability-20260820-model)

data/test.jsonl

@@ -0,0 +1,4 @@
+{"id":"ai-observability-2-1","input":"Prompt tokens doubled after a template change","label":"token-spike","context":"Token consumption increased beyond the cost and context baseline.","source":"trace-02","variant":"direct","synthetic":true}
+{"id":"ai-observability-3-2","input":"In an operations review, The retrieval tool returned a timeout exception","label":"tool-failure","context":"A required external tool failed during execution.","source":"trace-03","variant":"operations","synthetic":true}
+{"id":"ai-observability-4-3","input":"For an evaluation case, Grounded answer score dropped after deployment","label":"quality-regression","context":"Evaluation quality regressed relative to the release baseline.","source":"trace-04","variant":"evaluation","synthetic":true}
+{"id":"ai-observability-6-1","input":"Search calls failed with repeated connection errors","label":"tool-failure","context":"Dependency errors prevented the workflow from completing.","source":"trace-06","variant":"direct","synthetic":true}

data/train.jsonl

@@ -0,0 +1,14 @@
+{"id":"ai-observability-1-1","input":"Request latency rose above the service objective","label":"latency-regression","context":"End-to-end latency exceeded the approved percentile threshold.","source":"trace-01","variant":"direct","synthetic":true}
+{"id":"ai-observability-1-2","input":"In an operations review, Request latency rose above the service objective","label":"latency-regression","context":"End-to-end latency exceeded the approved percentile threshold.","source":"trace-01","variant":"operations","synthetic":true}
+{"id":"ai-observability-1-3","input":"For an evaluation case, Request latency rose above the service objective","label":"latency-regression","context":"End-to-end latency exceeded the approved percentile threshold.","source":"trace-01","variant":"evaluation","synthetic":true}
+{"id":"ai-observability-2-2","input":"In an operations review, Prompt tokens doubled after a template change","label":"token-spike","context":"Token consumption increased beyond the cost and context baseline.","source":"trace-02","variant":"operations","synthetic":true}
+{"id":"ai-observability-2-3","input":"For an evaluation case, Prompt tokens doubled after a template change","label":"token-spike","context":"Token consumption increased beyond the cost and context baseline.","source":"trace-02","variant":"evaluation","synthetic":true}
+{"id":"ai-observability-3-1","input":"The retrieval tool returned a timeout exception","label":"tool-failure","context":"A required external tool failed during execution.","source":"trace-03","variant":"direct","synthetic":true}
+{"id":"ai-observability-3-3","input":"For an evaluation case, The retrieval tool returned a timeout exception","label":"tool-failure","context":"A required external tool failed during execution.","source":"trace-03","variant":"evaluation","synthetic":true}
+{"id":"ai-observability-4-1","input":"Grounded answer score dropped after deployment","label":"quality-regression","context":"Evaluation quality regressed relative to the release baseline.","source":"trace-04","variant":"direct","synthetic":true}
+{"id":"ai-observability-4-2","input":"In an operations review, Grounded answer score dropped after deployment","label":"quality-regression","context":"Evaluation quality regressed relative to the release baseline.","source":"trace-04","variant":"operations","synthetic":true}
+{"id":"ai-observability-5-1","input":"The agent stayed within quality limits but became slower","label":"latency-regression","context":"Performance changed without a matching quality improvement.","source":"trace-05","variant":"direct","synthetic":true}
+{"id":"ai-observability-5-2","input":"In an operations review, The agent stayed within quality limits but became slower","label":"latency-regression","context":"Performance changed without a matching quality improvement.","source":"trace-05","variant":"operations","synthetic":true}
+{"id":"ai-observability-5-3","input":"For an evaluation case, The agent stayed within quality limits but became slower","label":"latency-regression","context":"Performance changed without a matching quality improvement.","source":"trace-05","variant":"evaluation","synthetic":true}
+{"id":"ai-observability-6-2","input":"In an operations review, Search calls failed with repeated connection errors","label":"tool-failure","context":"Dependency errors prevented the workflow from completing.","source":"trace-06","variant":"operations","synthetic":true}
+{"id":"ai-observability-6-3","input":"For an evaluation case, Search calls failed with repeated connection errors","label":"tool-failure","context":"Dependency errors prevented the workflow from completing.","source":"trace-06","variant":"evaluation","synthetic":true}

project.json

@@ -0,0 +1,52 @@
+{
+  "name": "Production AI Observability Monitor",
+  "problem": "Production AI teams need trace-level signals for latency, token growth, tool failures, and low-quality outputs.",
+  "domain": "ai-observability",
+  "architecture": "classifier",
+  "hugging_face_tasks": [
+    "text-classification",
+    "token-classification",
+    "summarization",
+    "zero-shot-classification"
+  ],
+  "recommended_stack": [
+    "OpenTelemetry GenAI semantic conventions",
+    "OpenTelemetry Collector for vendor-neutral ingestion",
+    "ClickHouse or PostgreSQL for trace analytics",
+    "Prometheus and Grafana for service-level metrics",
+    "MLflow for model and prompt version linkage",
+    "FastAPI for trace search and regression APIs"
+  ],
+  "real_world_data_sources": [
+    {
+      "name": "Prometheus HTTP API",
+      "url": "https://prometheus.demo.do.prometheus.io/api/v1/query?query=up",
+      "purpose": "Real time-series telemetry for anomaly pipelines"
+    },
+    {
+      "name": "GitHub Events API",
+      "url": "https://api.github.com/events?per_page=10",
+      "purpose": "Deployment-correlated public event metadata"
+    }
+  ],
+  "job_description_skills": [
+    "LLMOps observability and trace instrumentation",
+    "Prompt, model, dataset, and deployment lineage",
+    "SLOs, anomaly detection, and incident diagnostics",
+    "High-volume telemetry storage and aggregation",
+    "Evaluation-driven production monitoring"
+  ],
+  "impact_targets": [
+    "Ingest 1,000 synthetic traces/second without loss",
+    "Detect seeded latency and quality regressions with >= 0.90 precision",
+    "Link 100% of traces to model, prompt, and dataset versions",
+    "Generate a release health report in under 60 seconds"
+  ],
+  "baseline_evaluation": {
+    "test_examples": 4,
+    "accuracy": 1,
+    "synthetic_evaluation": true
+  },
+  "estimated_delivery": "8-12 weeks for one engineer",
+  "generated_baseline_is_production_ready": false
+}

sources.json

@@ -0,0 +1,12 @@
+[
+  {
+    "name": "Prometheus HTTP API",
+    "url": "https://prometheus.demo.do.prometheus.io/api/v1/query?query=up",
+    "purpose": "Real time-series telemetry for anomaly pipelines"
+  },
+  {
+    "name": "GitHub Events API",
+    "url": "https://api.github.com/events?per_page=10",
+    "purpose": "Deployment-correlated public event metadata"
+  }
+]