Cybersecurity Consultancy

Revolutionizing Compliance Document Reviews with Machine Learning

Machine Learning Compliance
AI-driven initiative
Prototype development
PCI-DSS standard
Embedding model

HATCHWORKS TEAM

Product Strategist, Senior Machine Learning Engineer, Senior Data Engineer

Overview

HatchWorks was engaged by a leading cybersecurity consultancy on a six-week strategic initiative to develop a prototype machine learning service. This service, leveraging the client’s data, was designed to predict the compliance status of a document. Once fully realized, this service promises to significantly reduce the time consultants invest in manually assessing customer compliance documents.

The Challenge

For the client, the manual review of customer compliance documentation was a significant bottleneck. Employing a vast team of consultants for this seemingly straightforward task of determining document compliance was not only resource-intensive but also fraught with inconsistencies and subjective interpretations.

The client faced challenges in:

Recognizing the potential of machine learning to address this challenge, and given their lack of in-house capabilities, the client sought HatchWorks’ expertise to help develop such a solution.

The Process

HatchWorks embarked on a six-week journey with a clear goal: to validate the feasibility of the envisioned service. The client’s role was pivotal in providing structured datasets from past projects to train the ML model.

Once the data was procured, it underwent a cleansing process and was structured for model training. Compliance requirements, specifically from the PCI-DSS standard, were ingested and transformed into optimized queries.

These queries enabled an embedding model to sift through vast documentation, extracting relevant contexts. This context was then processed by a fine-tuned language model, which determined compliance based on historical data. Throughout this process, HatchWorks maintained a collaborative approach, engaging with client stakeholders in virtual sessions. These sessions ensured data quality, model accuracy, and alignment with stakeholder needs, setting the stage for future projects.

The Outcome

By the project’s end, the language model showcased its capability to assess compliance documentation, pinpointing specific text references. Achieving this within six weeks underscored the potential of such a service.

The model’s performance, with 80% accuracy against test data, surpassed the initial 50% benchmark set by stakeholders. This success not only validated the project’s feasibility but also highlighted areas for future improvement, especially with more data.

“It has been a pleasure working with HatchWorks for almost two years now. They work seamlessly with my team, have overlapping time zones for efficient Agile development, and have a governance model to ensure success.

HatchWorks has become our partner for several successful ongoing projects.

I like their US-based Headquarters, and quality delivery teams working in Latin America. I plan to continue working with HatchWorks for our mutual journey and partnership.”

Cybersecurity Consultancy

Revolutionizing Compliance Document Reviews with Machine Learning

HATCHWORKS TEAM

Overview

The Challenge

Streamlining workflows for consistency.

Upholding quality standards.

Keeping consultants engaged and motivated.

The Process

The Outcome

-Chief Technology Officer @ Cybersecurity Consultancy