Case Study

AI agents automate merge processes

A leading German logistics provider is now benefiting from intelligent merge processes in the development of its software. This is made possible by an AI-based solution from the experts at Ki Reply.

#AI-powered development 
#Automated source code adjustment 
#Logistics sector


The challenge

Apply adjustments to the source code across the entire system - error-free and comfortably


The scenario

Seamless integration of blueprint changes

Our client, a leading German logistics provider, transports countless parcels every year. To enable it’s customers to easily track all these shipments online, a large number of services are required in the background - from parcel identification to providing an estimated delivery date. Each of these services is based on a blueprint that the logistics provider is constantly developing further. However, to ensure that tracking always works smoothly, these adjustments need to be seamlessly integrated into different copies of the source code, the repositories. Despite the many advanced guard rails, this merge was previously a time-consuming and error-prone process.

The solution


Transfer changes efficiently and reliably with AI

The integration of code adjustments into various repositories is now completely automated at our client. This is made possible by an AI-based solution from the software experts at Ki-Reply. Two intelligent agents carry out the following steps completely autonomously:

Creation of a new feature branch

The first AI agent creates a new development branch, a feature branch, in the corresponding repository. This is used to record the changes.

Transferring the changes from the main repository

Afterwards, the agent transfers the updates efficiently and reliably to the newly created feature branch.

Analyzing and resolving merge conflicts

When transferring the changes, the AI agent also pays attention to possible contradictions and problems, so-called merge conflicts, and applies automated solution strategies, which it also integrates into the feature branch.

Tests and subsequent adjustments

A second agent, which is specialized in testing software, now takes over. This agent uses the existing pipeline to check whether the system works properly with the changes. Any errors that occur are rectified directly.

Creation of a pull request for release

In the final step, the first AI agent takes over again and creates a pull request. This contains an explanation of the changes in natural language and an overview of the adjustments in the code and is used for final approval by the developer.

The concept in the background

Support along the entire development life cycle

The two AI-supported agents are managed by a multi-agent system, the KICODE Reply framework from Ki Reply. This supports the developers along the entire Software Development Life Cycle (SDLC). In this way, the experts at Ki Reply ensure that the new agent not only masters its task perfectly, but also delivers future-proof results. At the same time, it is compatible with other multi-agent systems and thus offers maximum flexibility.


The advantages

The full potential of generative AI

Thanks to the new AI-supported solution from Ki Reply, the logistics provider benefits from numerous advantages:


Accelerated software development


Less workload for the development teams


Tried and tested solutions and proven procedures


Fewer errors and contradictions in the code


More capacity for innovative projects

Ki Reply


Ki Reply is a service provider for AI-driven software development and supports companies in making software development more efficient and powerful with the help of artificial intelligence (AI) and machine learning. In addition to conventional software development, the experts at Ki Reply also use low-code platforms to realise and optimise business use cases more quickly. Companies benefit from time savings when testing IT solutions, a shorter time-to-market and high-quality software.