Fast On-Demand Team Setup and R&D for Telecom Audio Data Processing Center

Product Development, Backend Development, Frontend Development, Test Automation Services, Manual Testing Services, IT Consulting, Project Management, CI/CD Implementation & Modernization. DevOps Services, Early-Stage Innovation, R&D Services

A software company from USA, California working on building a telecom data processing solution.

A business intelligence platform for recording, storing, and analyzing telecommunications data.

The system collects audio content and its metadata, including call duration, time, source, etc., and provides an accessible repository for data storage. Metadata and the audio itself are conditionally divided and stored separately for easier access. The system also processes data from various sources at the same time.

Once the data is collected and filtered, the platform allows managers to transform and visualize it in a convenient and friendly format.

The platform is set up for unrestricted software integration so it can easily work with a variety of data coming from diverse sources.

This product was initially based on the customer’s already existing digital platform, which had massive amounts of documentation.

The products that make up the described system have a large codebase, complex integration logic, and multiple cases for working with various types of data in many combinations.

The problem was that the docs were either outdated or missing parts of the platform’s functionality. For instance, there was no clear description of the main use cases and the system’s architecture.

So the top challenges in development were to:

Development was launched urgently and on a tight schedule, with milestones connected to important commercial deliveries. Our team was called in to assist in R&D and streamline the development process to meet regular deadlines.

Another challenge was to set up proactive counter-measures against potential system failures caused by existing software integration issues, invalid server configuration problems, and third-party application compatibility challenges.

Developers needed to quickly grasp the workings of the system and be well-versed in the project context. To address the challenges, we deployed a dedicated team assigned with new developer onboarding and immersion.

They performed tasks that allowed us to quickly set up the infrastructure, assign appropriate roles, and launch solution development:
and tools:

Having laid this foundation, we created an efficient onboarding framework which reduced the time it takes new engineers to immerse in the project’s context from an enormous 3 months to merely 2 weeks.

Once we were done with groundwork, it was time to move on with development. To familiarize themselves with the project, new engineers were provided with a package of docs and detailed videos describing the system, its components, and their interaction with each other.

With a comprehensive set of prepared videos, diagrams, and guides, all team players were able to dive deep enough into the project and clearly understand the domain area of the product itself.

The client received a detailed multi-stage execution plan for platform refactoring and further development containing specific deadlines and simple descriptions of implementation techniques.

As part of the plan we focused our development on 3 key system components, each one of them responsible for its own functions:

This is the part that allows the platform to process data from multiple sources simultaneously.

It receives blocks of raw data, decodes them from their encoded formats, and then parses this data into distinct pieces of information. To allow processing data from multiple sources, we implemented multithreading features.

The processing component also creates requests and passes them on to the transcoding component for further data processing.

This is a media server that transcodes audio data and analyzes it to derive additional information. After transcoding, the analyzed content is saved and stored in the data repository.

This part of the system combines JavaScript and WebSocket to render, visualize, and play the recorded telecom audio content coupled with displayed metadata.

On the back-end, the management component generates the required web elements that are then displayed within the friendly interface to the users. All the database management tasks are handled by this component.

Under the leadership and active participation of JazzTeam’s lead architect, the plan was successfully put into action and the solution was delivered on time. The tasks were accomplished successfully and the speed of onboarding new engineers was quite impressive:

The average time an engineer took to immerse themselves into the project’s context was reduced from 3 months to 2 weeks.

This way we also eliminated the dependency on certain tech experts in explaining the project’s specifics to other team players. The prepared documentation provided an exhaustive description of the way the platform works.

It took us no more than 1 month to set up development within a distributed team with streamlined integration and delivery processes. We carried out extensive R&D and deployed the product within the required timeframe.

Thanks to our rigorous quality assurance and control methods, the telecom business intelligence platform was tested and proved to be flawless. We created a scalable solution architecture, so the client could easily develop new functionality in the future.

After our work on the documentation, any team could jump in and take over development from this point on. Our work on the project inspired the client to focus on their platform’s evolution and further development.

Backend Stack: Java 8, JSON, Spring, Gradle, GWT, JUnit, WebSocket.

Frontend Stack: GWT, HTML/CSS, JavaScript, D3.js, Canvas.

Infrastructure: Git, Bitbucket, Jira, Confluence, Jenkins, AWS, IntelliJ IDEA.

Frameworks: Apache Kafka, Docker, gRPC.

Test Automation libraries: Groovy + Java, Selenium WebDriver, Selenium Grid, ActiveMQ, TestNG, REST API.

Other libraries: Slf4j, Lombok.