VITAL4: SCALABLE AND DATA-DRIVEN
In an era of total interconnectivity, fallacious information has never run more rampant. We are forced to approach new collaborative relationships with a quiet skepticism, entertaining the hyperawareness needed to assuage risk. Vital4 nurtures a powerful data stream used to navigate unknown entities. They collect thousands of sources across the web, extracting and aggregating the data to form profiles on businesses and individuals.
Most organizations have some sort of background check policy in place to raise the red flags and seed out potential hires with a corrupt past. To automate this process, organizations engage with screening companies, sending over a list of names to run through and evaluate. Vital4 targets those pre-employment background screening companies, making them a B2B back-end partner in background screening processes.
Vital4 generates negative content relevant to the list of names their users provide. They offer their pre-employment or pre-engagement screening software through several engagement channels, including a search engine, integrated API, and web application for one-off searches.
Organizations receive comprehensive timelines which facilitate their fact-finding missions and eliminate misinformation. Vital4 filtrates watch lists and sanctions, political profiles, and negative media to engineer multidimensional online personas for their users.
Vital4 leverages a cloud-based SaaS solution to facilitate their globally accessible data search services. Their software allows participating parties to conduct more intensive preliminary investigations on customers, suppliers, contractors, partners, and volunteers for their contractual clients.
We first met Patrick Deeb, CTO at Vital4, at the Society of Corporate Compliance and Ethic’s annual conference in Las Vegas. Vital4 had just downsized their development team, putting them at a standstill. They intended on hiring two more resources to compensate for the programmers they lost, but they wanted to keep development progressing in the meantime.
Vital4 was amidst the development for a cutting-edge addition to their software suite, a PEP (Politically Exposed Person) product that formulates the political network associated with each query. The PEP product was a third to halfway through being completed to vision when Vital4 decided to bring us aboard.
Chetu engineers were tasked to enhance and enrich the preexisting code by examining the source media files based on each associated search term. Our second objective was to create a new PEP functionality to pinpoint the search term within the media files, putting the finishing touches on the PEP product Vital4 intended to add to their pre-employment background screening software.
We engineered a Console application through which the user can register and login before using its features. The login and registration can be accessed through their existing application as well. We dictated the role-based authentication and authorization, granting Admin users full authorization to update and enhance the data.
Next, we integrated Mechanism Rosette Text Analytics to extract the textual data from each media article and break it apart into a useable form, pulling out desired details like name, location, and relationship to the search term.
Web Application Modules
Our project parameters were outlined as follows, dictating a two-prong approach. The proposed system required the following functionalities:
1. Enhance and enrich the existing PEP Profile.
- Search for the matching media articles in the source database based on the keyword provided.
- Store the reference of the matching records in new database table.
- Loop through each of the existing PEP Profiles and use the Rosset Name Matching tool to find the matching name.
- If a match is found, use the Rosette Entity matching tool will match the different entity of the PEP profile holder.
- Enrich the findings by updating the entity into existing PEP Profile database fields.
2. Create New PEP Profile:
- Use the newly stored media articles from Step 1.
- Identify new individuals who hold positions of authority.
- Identify the individuals name associated with each keyword.
- Compare with existing PEP Profile to check if the individual has an existing PEP Profile.
- If the system finds an individual’s name and position of authority, then it should further compare the positions (Country and State/Region). If no match is found, then discard that article and move into the next article /record
- If PEP Profile does not exist in the system, or if it exists and the position and the location does not match, the system should create a new PEP profile.
We engaged with five different modules during our project with Vital4. These modules included a Service Module, PEP Profile Enrichment Module, PEP Profile Creation Module, Rosset Text Analytics, and PEP Database.
Service Module: This module will contain the basic logic required to process media files from the starting /entry point.
Profile Enrichment: This module will be responsible for capturing the media-relevant files based on the keyword defined and use the Rosette Text Analytics to match the name and entity and update the existing profile with different entity.
PEP Profile Creation: This Module will be responsible for identifying the media articles that are related with a particular position and name and use that media to create the basic PEP Profile.
Rosette Text Analytics: This Module already exists and is running on production above both module will use this to get the results of Name and Entity Matching.
Database: Database where Profile Enrichment and PEP Profile Creation module will update/create the new profile data.
If we zoom in on the PEP product, we can dissect it into three layers: the presentation layer, a business layer, and the data access layer. We built this hierarchy to forge sustainability—the component-level design supports scalability and incremental enhancement.
For the development framework, we went with Visual Studio as the IDE, C# as the programming language, and SQL as the Database. The console application is compatible with Windows operating system, and supports a variety of browsers including Chrome, IE, Firefox, and Safari.
ANALYZING THE RESULTS
During our post-project interview with Vital4, they discussed the power of our communication structure. Our project manager, “asked all the right questions,” regarding the development, and took the necessary steps to understand the client’s business context. Unfortunately, the third-party plugin, Rosette Text Analytics, that Vital4 included in the original roadmap proved faulty and failed to return the results they anticipated. In terms of execution, the Chetu fulfilled all the parameters and ultimately turned over a very functional final source code. An alternative third-party plugin, would need to be replaced and reintegrated into the code we prepared.
Compared to some of our other engagements, our relationship was fairly short-term. They experienced a momentary lapse in development resources, and we were able to fill the gap. Our relationship concluded after Vital4 on-boarded additional developers, replacing the resources they lost. This is one of the most compelling facets of the Chetu model—total scalability.
Our clients scale our involvement to their needs, pressing pause or play to parallel their internal agenda. The scale tips in both directions; sometimes projects are a one-off and other times our engagements continue for years, transitioning from outsourced development to a long-term, collaborative partnership. Perhaps, later down the line, we will see Vital4 again.