Jobs API Reference

This page documents the gRPC services that back job submission and lifecycle management in Michelangelo AI. Two services are defined in proto/api/v2/:

• RayJobService — create, inspect, and delete Ray training or batch inference jobs.
• SparkJobService — create, inspect, and delete Spark ETL and batch processing jobs.

New to this reference? See How to Read the gRPC API Reference for the shared CRUD-plus-list pattern, the Required column, and how undocumented fields are shown. For how jobs run end-to-end, see Michelangelo AI Jobs.

---

RayJobService

RayJob Service defines the RayJob related methods, such as CRUD and list.

Proto source: proto/api/v2/ray_job_svc.proto

CreateRayJob

CreateRayJob(CreateRayJobRequest) → CreateRayJobResponse

Create a new RayJob with the given spec.

Request fields

Field	Type	Required	Description
ray_job	RayJob	Yes	The metadata and spec of the RayJob to be created.
create_options	k8s.io.apimachinery.pkg.apis.meta.v1.CreateOptions	No	—

Response fields

Field	Type	Description
ray_job	RayJob	The created RayJob.

---

GetRayJob

GetRayJob(GetRayJobRequest) → GetRayJobResponse

Get the specified RayJob.

Request fields

Field	Type	Required	Description
name	string	Yes	Name of the RayJob.
namespace	string	Yes	Object name and auth scope.
get_options	k8s.io.apimachinery.pkg.apis.meta.v1.GetOptions	No	—

Response fields

Field	Type	Description
ray_job	RayJob	The requested RayJob.

---

UpdateRayJob

UpdateRayJob(UpdateRayJobRequest) → UpdateRayJobResponse

Update a RayJob with the given spec.

Request fields

Field	Type	Required	Description
ray_job	RayJob	Yes	The metadata and spec of the RayJob to be updated.
update_options	k8s.io.apimachinery.pkg.apis.meta.v1.UpdateOptions	No	—

Response fields

Field	Type	Description
ray_job	RayJob	The updated RayJob.

---

DeleteRayJob

DeleteRayJob(DeleteRayJobRequest) → DeleteRayJobResponse

Delete a RayJob.

Request fields

Field	Type	Required	Description
name	string	Yes	Name of the RayJob.
namespace	string	Yes	Object name and auth scope.
delete_options	k8s.io.apimachinery.pkg.apis.meta.v1.DeleteOptions	No	—

Response fields

The response message carries no fields.

---

DeleteRayJobCollection

DeleteRayJobCollection(DeleteRayJobCollectionRequest) → DeleteRayJobCollectionResponse

Delete collection of RayJob.

Request fields

Field	Type	Required	Description
namespace	string	Yes	Object name and auth scope.
delete_options	k8s.io.apimachinery.pkg.apis.meta.v1.DeleteOptions	No	—
list_options	k8s.io.apimachinery.pkg.apis.meta.v1.ListOptions	No	—

Response fields

The response message carries no fields.

---

ListRayJob

ListRayJob(ListRayJobRequest) → ListRayJobResponse

List objects of type RayJob.

Note: Watch and list across all namespaces are not supported.

Request fields

Field	Type	Required	Description
namespace	string	Yes	—
list_options	k8s.io.apimachinery.pkg.apis.meta.v1.ListOptions	No	—
list_options_ext	michelangelo.api.ListOptionsExt	No	—

Response fields

Field	Type	Description
ray_job_list	RayJobList	—

---

SparkJobService

SparkJob Service defines the SparkJob related methods, such as CRUD and list.

Proto source: proto/api/v2/spark_job_svc.proto

CreateSparkJob

CreateSparkJob(CreateSparkJobRequest) → CreateSparkJobResponse

Create a new SparkJob with the given spec.

Request fields

Field	Type	Required	Description
spark_job	SparkJob	Yes	The metadata and spec of the SparkJob to be created.
create_options	k8s.io.apimachinery.pkg.apis.meta.v1.CreateOptions	No	—

Response fields

Field	Type	Description
spark_job	SparkJob	The created SparkJob.

---

GetSparkJob

GetSparkJob(GetSparkJobRequest) → GetSparkJobResponse

Get the specified SparkJob.

Request fields

Field	Type	Required	Description
name	string	Yes	Name of the SparkJob.
namespace	string	Yes	Object name and auth scope.
get_options	k8s.io.apimachinery.pkg.apis.meta.v1.GetOptions	No	—

Response fields

Field	Type	Description
spark_job	SparkJob	The requested SparkJob.

---

UpdateSparkJob

UpdateSparkJob(UpdateSparkJobRequest) → UpdateSparkJobResponse

Update a SparkJob with the given spec.

Request fields

Field	Type	Required	Description
spark_job	SparkJob	Yes	The metadata and spec of the SparkJob to be updated.
update_options	k8s.io.apimachinery.pkg.apis.meta.v1.UpdateOptions	No	—

Response fields

Field	Type	Description
spark_job	SparkJob	The updated SparkJob.

---

DeleteSparkJob

DeleteSparkJob(DeleteSparkJobRequest) → DeleteSparkJobResponse

Delete a SparkJob.

Request fields

Field	Type	Required	Description
name	string	Yes	Name of the SparkJob.
namespace	string	Yes	Object name and auth scope.
delete_options	k8s.io.apimachinery.pkg.apis.meta.v1.DeleteOptions	No	—

Response fields

The response message carries no fields.

---

DeleteSparkJobCollection

DeleteSparkJobCollection(DeleteSparkJobCollectionRequest) → DeleteSparkJobCollectionResponse

Delete collection of SparkJob.

Request fields

Field	Type	Required	Description
namespace	string	Yes	Object name and auth scope.
delete_options	k8s.io.apimachinery.pkg.apis.meta.v1.DeleteOptions	No	—
list_options	k8s.io.apimachinery.pkg.apis.meta.v1.ListOptions	No	—

Response fields

The response message carries no fields.

---

ListSparkJob

ListSparkJob(ListSparkJobRequest) → ListSparkJobResponse

List objects of type SparkJob.

Note: Watch and list across all namespaces are not supported.

Request fields

Field	Type	Required	Description
namespace	string	Yes	—
list_options	k8s.io.apimachinery.pkg.apis.meta.v1.ListOptions	No	—
list_options_ext	michelangelo.api.ListOptionsExt	No	—

Response fields

Field	Type	Description
spark_job_list	SparkJobList	—

---

Next Steps

• Michelangelo AI Jobs: Core concepts, architecture, and how Ray/Spark jobs run end-to-end
• Run a Pipeline on a Compute Cluster: Submit and monitor a Uniflow pipeline on a registered cluster