Workflow

Getting Set Up

To use the extraction API, you will need to go through an onboarding process that ensures that set up requirements are met. Contact Emtelligent support at support@emtelligent.com to initiate onboarding steps (e.g. account creation, access to pipelines)

Onboarding requires setup by both the customer and emtelligent. This page outlines the customer vs. emtelligent setup requirements.

Emtelligent-provided Requirements

  • Access to this documentation
  • An API key for the Extract API
  • Designated input and output S3 buckets

Storage Requirements

  • Emtelligent provides bucket names during onboarding. If you wish to use different bucket names, please contact emtelligent support.
  • Buckets should be used exclusively for extraction workflows.
  • The customer is solely responsible for managing access control for these buckets.
  • Within these buckets you can use prefixes to organize files.

Customer-provided Requirements

You will need to provide the following:

  • AWS Account with S3 Access
  • AWS S3 buckets with appropriate access permissions
  • Data set for processing

AWS Account & S3 Access

You will need to have the following:

  • Active AWS account with ability to create and manage S3 buckets
  • Permissions to create IAM policies and generate access credentials
  • Basic familiarity with S3 bucket operations and IAM permissions

AWS S3 Buckets

You must set up two S3 buckets named as agreed with emtelligent:

  • An Input bucket: where you will upload documents to be processed.
  • An Output bucket: where extraction results will be stored.

You must have sufficient privileges to read/write to these buckets. Additionally, you should have permissions to issue AWS STS Tokens that you will provide in the job request, giving emtelligent temporary access to ingest from and write to the respective buckets.

S3 IAM Permission setup

Input Bucket Access Permissions

  • Allowed Actions: ListObjects, ListObjectsV2, GetObject
  • Scope: Bucket or appropriate subset of prefixes covering all manifest object keys
  • Credential Type: AWS IAM Access/Secret keys or STS AssumeRole
  • Expiration time (STS): As appropriate for dataset size or up to max allowed

Output Bucket Access Permissions

  • Allowed Actions: ListObjects, ListObjectsV2, PutObject
  • Scope: Bucket resource
  • Credential Type: AWS IAM Access/Secret keys or STS AssumeRole
  • Expiration time (STS): As appropriate for dataset size or up to max allowed

Data Set (with optional metadata)

You will need to:

  • Identify a set of PDF documents ready for processing
  • Organize any metadata you wish to associate with those PDFs. See Preparing Files for Processing.
  • Prepare a file manifest for each job

Checklist

Have you:

  • [ ] Received API key from emtelligent
  • [ ] Confirmed that emtelligent has set up your account
  • [ ] Received designated bucket names
  • [ ] Set up input S3 bucket with correct IAM permissions
  • [ ] Set up output S3 bucket with correct IAM permissions
  • [ ] Tested your AWS credentials (S3 credentials with sufficient permissions to generate STS credentials)
  • [ ] Checked that you have access to the input and output buckets

You are now ready to start extracting data by submitting a job. For a step-by-step guide on how to submit a job, see Getting Started Tutorial.