Custom Software

Safeguarding Data with AI-Driven Employee Monitoring

AcumenAI is a tool designed to monitor employee communications and prevent data breaches across various systems.

TIMELINE: 1 YEAR 1 MONTH 
COUNTRY: ISRAEL
TEAM: 2 DEVELOPERS

About

AcumenAI uses advanced file scanning and real-time monitoring to help businesses detect and prevent data breaches across employee communications.

Result in Short: This system enables the client to monitor and analyse employee files for potential data breaches across different systems.

The Client & The App

AcumenAI is a project initiated by a client from Israel with the goal of developing an AI-powered tool to monitor employee communications for potential data breaches.

Initially envisioned as an application leveraging ChatGPT to monitor employee chat conversations, the project evolved into a broader solution that includes a robust file scanning system designed to identify confidential information across multiple file types and storage locations.

The primary objective of the app is to provide companies with a security solution capable of identifying data leak threats, especially within environments where employee communication and document sharing are prone to risks of exposing sensitive information.

This comprehensive tool includes both file scanning and real-time chat monitoring, with AI-powered capabilities to detect confidential data like personal information, credit card numbers, and attempts to manipulate AI models.

The project relies heavily on close collaboration with the client. This includes understanding their vision, managing their expectations, and effectively communicating technical details. The collaborative approach ensures that the solution meets the client's specific needs and addresses their evolving security concerns.

Did you know?

The use of AI in front-end development showcases significant time savings and increased efficiency in software development.

Our Role and Involvement

They’ve found our developer from Upwork. Initially brought in for DevOps support to set up the infrastructure for the application using Terraform.  Our involvement focused on stabilising and enhancing the app's foundation, but it expanded to full-stack development, encompassing infrastructure setup as well as front-end and back-end development.

Additionally, we integrated a Machine Learning Model created by a separate team, ensuring it functioned seamlessly within the app.

We provided guidance and direction to the client, particularly in terms of roadmap and feature prioritisation.

Results

Deployment and Sources

Successful deployment of a functional file management system which scans and analyses files from various sources (currently supporting SFTP and local systems, with Google Drive and other cloud services on the roadmap).

Categorisation and Sensitive Data

The system categorises files and identifies sensitive data like names, addresses, credit card numbers, etc. It's multi-tenant, meaning it can handle multiple clients' data within a single database.

Scalability and Security

The back-end was built for scalability, using cloud solutions to handle large data volumes and optimise performance. Prioritizing security, we implemented strict encryption protocols and regular audits. A custom algorithm detects sensitive information in real-time based on predefined criteria.

Chat Application Filtering Component

This component will act as a filter for chat applications within company networks, flagging potentially problematic content to administrators. It builds upon a previous, now-scrapped project involving ChatGPT and data privacy.

KPI

System capable of analysing up to 700 files per second.

Significant time savings in front-end development using AI-assisted coding tools.

The Process

The development process is characterised by agility and iteration. The client's rapidly changing needs and tendency to introduce new requirements necessitate a flexible approach.

The development process involves not just building new features but also troubleshooting, optimising, and refining existing code. This includes addressing performance issues like the system initially being "too fast" and overloading client servers, requiring the implementation of rate limiting. It also involves refactoring and improving upon the work of the previous developer.

While the Machine learning model was developed by a separate team, our developer was responsible for integrating it with the file scanning system and ensuring smooth data flow.

Initial Engagement

DevOps support for application infrastructure.

Expanding Scope

Code review, bug fixing, and auditing.

Taking Ownership

Leading development of the file management system (front-end, back-end, infrastructure).

Iterative Development

Working closely with the client to refine features and prioritise the roadmap.

Launch and Ongoing Support

Deploying the system and continuing to add features and integrations.

Let's win your market together!

Tell us more about your application

Contact us to discuss your app idea and possibilities. We’ll advise you on the best solution and estimate the project. If you have any questions – we’ll provide you with answers.

Let's talk!

Schedule a call with Mark,
our Technical Solutions Manager

Write a message

mark.cameron@teacode.io

Schedule a meeting

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form.