slug: seldon provider: Seldon generated_by: planning/capability-mapping/scripts/classify_capabilities.py model: claude-opus-5 frame: - Software & Technology min_confidence: 0.7 capability_model: source: https://github.com/vincentmakes/turbo-ea-capabilities license: CC-BY-4.0 attribution: Turbo EA Capabilities by Vincent Verdet — Turbo EA, https://github.com/vincentmakes/turbo-ea-capabilities, CC BY 4.0 notice: NOTICE edge_count: 14 edges: - tag: PermissionManagementService spec_file: seldon-permissionmanagementservice-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.88 evidence: '''Create an authorization policy.'', ''List all permissions associated with the given users and groups.'', GET /iam/users' reason: Users, groups and authorization policies under /iam paths are plainly identity and access management. - tag: ModelMetadataService spec_file: seldon-modelmetadataservice-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.85 evidence: '''Create a Model Metadata entry'', ''List Runtime Metadata for all deployments associated with a model''; schemas v1PredictionSchema, v1ArtifactType' reason: A model metadata registry describing ML models, artefacts and prediction schemas is core AI/ML model lifecycle (MLOps) management. - tag: Models spec_file: seldon-models-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.85 evidence: POST /namespaces/{namespace}/models 'Create model'; schemas SeldonModel, ModelSpec, ExplainerSpec reason: CRUD over machine-learning model resources (with explainer specs) on an MLOps platform realises AI/ML model lifecycle management. - tag: SeldonDeployments spec_file: seldon-seldondeployments-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.82 evidence: '''Create seldon deployment'', ''Validate seldon deployment''; schemas SeldonDeployment, PredictorSpec, PredictiveUnitImplementation' reason: Deploying and validating ML model serving deployments is model deployment within the AI/ML (MLOps) lifecycle. - tag: DriftDetector spec_file: seldon-driftdetector-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.8 evidence: POST /namespaces/{namespace}/pipelines/{name}/monitor/drift-detector CreateDriftDetectorSeldonPipeline reason: Create/read/delete drift detectors attached to deployed models and pipelines with schemas like DetectorConfiguration and DetectorStatus — model drift monitoring, a core MLOps/AI model lifecycle activity. - tag: Explain spec_file: seldon-explain-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.8 evidence: POST /namespaces/{namespace}/seldondeployments/{name}/explain ExplainSeldonDeployment Explain seldon deployment reason: Model explainability for deployed ML models is squarely AI/ML model lifecycle and responsible-AI tooling (BC-610.60); vendor description states the platform 'explain[s] machine learning models at scale'. - tag: Inference spec_file: seldon-inference-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.78 evidence: POST /v2/models/{model_name}/infer model-inference Model Inference; schemas InferenceRequest, InferenceResponse reason: Serving predictions from deployed ML models is the runtime execution leg of the AI/ML model lifecycle (BC-610.60 Artificial Intelligence Management). - tag: Monitor spec_file: seldon-monitor-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.78 evidence: '''Seldon deployment feature distributions'', ''Seldon pipeline feature statistics''; schemas FeatureDistribution, FeatureStats' reason: Feature distribution and statistics monitoring of deployed ML models/pipelines is model monitoring within the MLOps lifecycle, not generic service observability. - tag: OutlierDetector spec_file: seldon-outlierdetector-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.78 evidence: '''Create outlier detector seldon deployment''; schemas DetectorConfiguration, DetectorStatus' reason: Lifecycle management of outlier detectors attached to model deployments is ML model monitoring / responsible-AI tooling, part of AI/ML management. - tag: Pipelines spec_file: seldon-pipelines-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.78 evidence: '''Create pipeline'', ''Create pipeline explainer'', ''List pipeline models''; schemas SeldonPipeline, PipelineStep, ModelStatus' reason: Inference pipelines composed of ML models and explainers are ML serving artefacts; their lifecycle management falls under AI/ML (MLOps) management. - tag: MetricsServer spec_file: seldon-metricsserver-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.75 evidence: POST /namespaces/{namespace}/seldondeployments/{name}/monitor/metrics-server — 'Create metrics server seldon deployment'; schemas DetectorConfiguration, DetectorStatus reason: Operations deploy and manage drift/metrics detectors attached to deployed ML models — monitoring within the ML model lifecycle (MLOps), which is AI/ML management rather than generic IT observability. - tag: Experiments spec_file: seldon-experiments-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.7 evidence: POST /namespaces/{namespace}/pipelines/{name}/experiment/canary CreateCanaryExperiment; PUT .../shadow PromoteShadow reason: Canary/shadow experiments over ML pipelines are part of the model deployment lifecycle on an MLOps platform, which BC-610.60 explicitly names (AI/ML model lifecycle, MLOps). Progressive-deployment reading (BC-4210.40) is also plausible, hence moderate confidence. - tag: LoadtestJobs spec_file: seldon-loadtestjobs-api-openapi.yml capability_id: BC-4220.50 capability_id_l1: BC-4220 capability_name: Capacity & Performance Management confidence: 0.7 evidence: POST /namespaces/{namespace}/seldondeployments/{name}/loadtestjobs CreateLoadtestSeldonDeployment Create loadtest seldon deployment reason: Creating and managing load-test jobs against deployed services is performance testing, named in BC-4220.50 Capacity & Performance Management. - tag: Predict spec_file: seldon-predict-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.7 evidence: POST /namespaces/{namespace}/seldondeployments/{name}/predict PredictSeldonDeployment 'Predict seldon deployment' reason: Inference/prediction invocation against deployed models is the runtime of the ML model lifecycle; no other candidate capability fits, though the surface is thin (only an Error schema).