All countries

Price: 0

Number of applications: 1

Decision acceptance deadline

26.01.26 (inclusive)

Form of award

Monetary

Product status

Idea

Task type

ICT tasks

Сфера применения

Media sphere

Область задачи

Other technological solutions

Type of product

Software/ IS

Problem description

Modern software product development is characterized by high competition and the need to quickly bring new features to market. The standard production cycle includes development, code review, and testing. An analysis of current processes has shown that the code review stage is one of the most resource-intensive: in some cases, it takes up to 30-40% of the total task development time, reducing the speed of implementation of new functionality and increasing the burden on engineering teams. Existing code review practices require significant specialist involvement, which makes it difficult to scale development and leads to increased project completion times. This creates an objective need to automate the code review process without loss of quality.

Expected effect

The project implementation will allow: – To automate the code review process based on modern artificial intelligence technologies. – Reduce the labor costs of engineering teams by 30-40%. – Increase the release rate of new functionality without expanding staff. – Improve the quality of code verification by following common rules and standards. – Increase the predictability and transparency of the development process.

Full name of responsible person

Bobylev Oleg

Purpose and description of task (project)

As part of the project, it is planned to develop a tool that will automatically: 1. Track the opening of new Pull Requests. 2. Analyze the compliance of the provided code with: – The requirements of the terms of reference, – Internal standards and development agreements, – The correctness of using third–party libraries and their versions, - The structure and quality of the implemented logic. 3. Generate recommendations and reports in a format suitable for future use by engineers. Modern language models have sufficient ability to contextually analyze code when providing source data: task descriptions, specifications, and internal regulations. This makes it possible to significantly increase efficiency and reduce dependence on the human factor when conducting a code review. The project is aimed at increasing the technological maturity of development processes, optimizing the use of engineering resources and ensuring a steady increase in the production rate of software solutions.

Note