Box.com Discovery Bates Namer: Cloud Integration
Migration from Desktop Tool to Serverless AWS Workflow

Project Sponsor:
Deepak Budwani, Chief Financial and Administrative Officer – Office of the Public Defender, Santa Barbara County

Project Managers and Primary Project Contact:
AJ Voisan

Project Team and Subject Matter Experts:
Angie Stokke, Sarah Rothschild, Luis Ramirez, and Jalena Jones

Description

As the Office of the Public Defender continues its transition to a fully paperless case management system, we have adopted Box.com as our secure, CJIS- and HIPAA-compliant cloud content management solution. This platform allows for efficient digital file storage, internal collaboration, workflow automation, and integration with external systems like our case management software, eDefender.

With data ingestion exceeding 5–6 terabytes of electronically stored information (ESI) each month, automating file organization and metadata handling has become critical. One key project this term focuses on automating the discovery file workflow—specifically, parsing uploaded PDFs for Bates stamps, renaming those files, and transferring them to the appropriate locations in Box based on their case number.

In a prior phase of this project, a standalone desktop application was developed to assist with Bates parsing and file renaming. However, the manual workflow—downloading files, running the tool locally, and re-uploading renamed documents into Box—was inefficient and time-consuming. To address this, our team migrated the solution into a fully serverless cloud environment using AWS Lambda.

Now, when users upload discovery materials into a designated folder in Box, the system is triggered via webhook. AWS Lambda functions validate the input, extract Bates stamp ranges from the file content, rename the files accordingly, and move them to the appropriate destination folders. The system handles errors gracefully, distinguishing between soft and critical failures, and notifies the uploader via email if any issues occur.

This automation ensures consistent file naming and precise placement within the Box folder structure, dramatically reducing the manual effort and potential for human error in discovery processing. By eliminating the need to download, locally process, and re-upload files, the system not only saves time and computing resources, but also reduces risk by minimizing unnecessary file transfers between systems, thereby removing vulnerability of potential man in the middle attacks. We migrated the original desktop application into a fully serverless, cloud-native pipeline powered by AWS Lambda—delivering an end-to-end solution that is faster, more scalable, and inherently more secure. With built-in error detection, graceful fallback handling, and real-time email notifications, the system is both robust and user-friendly. Ultimately, this project streamlines a critical workflow, improves operational safety and reliability, and lays the foundation for future enhancements in automated document handling and metadata intelligence.