Software Development

End-To-End Case Study

Geo location

US (South)

Engagement

ETM

Service

Software Development

Challenges

Our client wanted to incorporate agile software development practices, including software application development, production application support, database development and administration, and DevOps support. They wanted this to all be done at the same time as they onboarded a new natural language processing (NLP) app and while they migrated to a new database. The US-based market researcher was on the lookout for a software development company that would match their ambition, innovation, and drive.

Updating systems

Overhauling processes

Migrating data

Tracking everything

The client required a new application to update their writing process
They also needed a new database to be developed and monitored
Extensive production application support was essential as they integrated new systems and migrated everything at once
They wanted to integrate DevOps support to ensure nothing was overlooked

Our Solution

As a software development service provider, we knew a project as complex as this one would require our most advanced services. To provide the best possible machine learning and cloud migration solutions, we built software engineering into the heart of everything we did. We assembled a team who could provide all the software application development services the client required and utilized the proper technology to ensure the whole project was completed on time and under budget. Our projects were all individually successful because we started with a foundation in software engineering.

Solution Impact

Time & budget constraints followed

Our software development solutions ensured resources were brought online and onboarded on time and within the budget

Customized for seamless integration

We evaluated and recommended Jira Service Desk as the ticket management tool

5%

Reduction

In the client’s post-deployment defect density

Our Approach

Before we started the more high-profile projects for this client, we looked at the development strategies that would be needed to underpin the projects. We defined the support delivery processes and recommended a ticket management tool setup that would support these processes. We then proceeded to utilize this as we undertook the main projects.

Omni-channel Retail – Change the rules of the game with holistic order management solutions

Read Blog

Building software engineering into the project

Our client recognized that building a firm base of software engineering into the project would make it much more successful than trying to piece several smaller custom software product development projects together at the end. That is why we began by defining the support delivery processes. We took care to align those processes with the client’s end business goals. We recommended a ticket management tool, Jira Service Desk. We then helped configure and customize the client’s version of the service desk to help manage tickets raised against software issues. When issues were raised, the client and our team were able to address them quickly, tracking and recording successes and solutions.

Increasing writing quality and efficiency with automation

Our award-winning software design service involved using natural language generation (NLG) to help the client’s survey writers generate false or incorrect answers to their survey questions. We built an app that used Terraform scripting to automatically generate responses to survey questions. Writers could then select from these options. If the writers chose an option, the program would learn that answer was acceptable. The writer could also generate more options, which would tell the program that the options didn’t work for the writer. The program would use this feedback to inform and improve future options. This constant learning and evolution means the program continued to increase the writers’ productivity. The auto-generated questions even increased the productivity of new writers, as they no longer had to rely on more experienced writers for training. This project was so successful, in fact, that the client asked us to discover new ways to integrate automation into other parts of their business.

Speeding up database migration with data testing

Our client wanted to migrate to a new cloud-based database while we also automated content generation. We provided AWS cloud migration support that included software product development and enhancements using Java and Angular, with product support from Oracle PL/SQL. We also used Oracle for database administration. When we began migrating data, we built a testing suite using Manual, Selenium, and FitNesse that allowed us to validate selected test data during the migration. That way, we could verify that data could be migrated accurately, without the need to test every single file and piece of data. This sped up the process of migration significantly. The testing suite also highlighted failed files and generated reports in Excel and in the FitNesse wiki. That gave us the necessary clarity and allowed us to manage the quality of the data during the migration without slowing the whole process down.

Business Impact

Because they valued our expertise and ambition as a custom software development firm, our client enjoyed numerous successes. Our NLP services helped reduce the time it took to generate questions, from 10 minutes to 6.7. Our cloud migration support allowed the client to validate 500,000 files at a rate of 1,000 files per minute. These successes were all because of the software engineering we instilled at the beginning. Those practices led to their own successes as well, including: