Create job

Authorizations

x-pd-api-key

string

header

required

Headers

x-pd-external-trace-id

string

The trace ID you set in your system to trace this request. It can be up to 128 characters in length. If the request fails, you can provide it to the Powerdrill team to help with troubleshooting.

Body

application/json

session_id

string

required

The session ID.

To check sessions you created, call GET /v2/team/sessions.

question

string

required

The question you want Powerdrill to answer.

user_id

string

required

Your user ID, which uniquely identifies you within your team. To obtain your ID:

If you're the team admin, refer to Check user information.
If you're a system or virtual user, ask your team admin to check your user ID by referring to Check user information.

stream

boolean

default:

false

Whether to use stream mode. If set to true, Powerdrill will send real-time updates to the client, delivering continuous data as it becomes available. If set to false, Powerdrill will return the full response only once the entire answer is ready.

If not specified, the default value false will be used.

For details about how to understand streaming responses, see "Content description" in Streaming.

dataset_id

string

The ID of the dataset to attach to the job.

To check the datasets you have access to, call GET /v2/team/datasets.

datasource_ids

string[]

Specifies the IDs of the data sources to use in the conversation, rather than the entire dataset. You can specify up to 1,000 data sources. This parameter is effective only when dataset_id is specified.

output_language

enum<string>

default:

AUTO

The language in which the output is generated. For example, if set to EN, the output will be in English. If not specified, the session's output_language setting will be used. Possible values are:

AUTO: adaptive recognition
EN: English
ES: Spanish
AR: Arabic
PT: Portuguese
ID: Indonesian
JA: Japanese
RU: Russian
HI: Hindi
FR: French
DE: German
VI: Vietnamese
TR: Turkish
PL: Polish
IT: Italian
KO: Korean
ZH-CN: Simplified Chinese
ZH-TW: Traditional Chinese

Available options:

AUTO,

EN,

ES,

AR,

PT,

ID,

JA,

RU,

HI,

FR,

DE,

VI,

TR,

PL,

IT,

KO,

ZH-CN,

ZH-TW

job_mode

enum<string>

default:

AUTO

Job mode. Possible values are:

AUTO: Powerdrill automatically detects your intent and selects the appropriate mode for data analysis or information retrieval.
DATA_ANALYSIS: Powerdrill focuses solely on data analysis.

If not specified, the session's job_mode setting will be used.

Available options:

AUTO,

DATA_ANALYTICS

Response

200

application/json

code

integer

required

Status code. 0 indicates that the operation is successful. Otherwise, the operation fails. For error troubleshooting, refer to Error Codes.

data

object

required

A job object.

data.job_id

string

required

The job ID, which uniquely identifies the job in the session.

data.blocks

object[]

required

A list of answer blocks that make up the entire answer.

data.blocks.type

enum<string>

required

The content type of the answer block. Possible values are:

MESSAGE: The content is a piece of text.
CODE: The content a code snippet.
TABLE: The content represents a table.
IMAGE: The content represents an image.
SOURCE: The content represents a reference source of the answer block.
QUESTIONS: Suggested questions generated by Powerdrill to help guide your follow-up exploration of the data.

Available options:

MESSAGE,

CODE,

TABLE,

SOURCES,

QUESTIONS

data.blocks.content

string

required

The block content, which varies with the value of type:

When type is MESSAGE, the content is a piece of text.
When type is CODE, the content a code snippet in Markdown format.
When type is TABLE, the content represents a table, consisting of:
- name: The .csv file name.
- url: The S3 key or URL to the file.
expires_at: The expiration time for url. To save the table for future use, make sure to download it before it expires.
When type is IMAGE, the content represents an image, consisting of:
- name: The image name.
- url: The S3 key or URL to the image.
expires_at: The expiration time for url. To save the image for future use, make sure to download it before it expires.
When type is SOURCE, the content represents a reference source of the answer block, including:
- source: The file name of the data source.
- datasourceId: The ID of the data source.
- dataset_id: The ID of the dataset.
file_type: The name extension of the data source file.
When type is QUESTIONS, the content is suggested questions generated by Powerdrill to help guide your follow-up exploration of the data.

data.blocks.group_id

string

required

The ID of the group containing the answer block.

data.blocks.group_name

string

required

The name of the group containing the answer block.

data.blocks.stage

enum<string>

required

There are two phases when Powerdrill generates an answer: Analyze and Respond. Answer blocks in the Analyze phase are not part of the final answer; they represent the output of the analysis process and are used to help you understand how the answer is generated. Answer blocks in the Respond phase are the final answer generated by Powerdrill in response to your question.

Available options:

Analyze,

Respond

Overview

Datasets

Data sources

Files

Sessions

Jobs

Authorizations

Headers

Body

Response