Price: 1500000
Number of applications: 2
25.12.25 (inclusive)
tenge
Idea
ICT tasks
Robotics
Intelligent control systems
Software/ IS
Preparation of project documentation for software development takes considerable time and requires the participation of analysts, project managers and developers. User stories, acceptance criteria, and functional requirements are generated manually, which leads to high labor costs, quality heterogeneity, and the risk of errors. The lack of a single standardized process leads to dependence on the qualifications of individual specialists, reduces the speed of project launch and increases the cost of training. In addition, the inability to quickly update documentation when requirements change leads to the accumulation of technical debt and reduced transparency of development. A tool is needed that can automatically analyze input descriptions, structure them, generate full-fledged artifacts of project documentation, and provide a single format for all company projects.
The implementation of the ZEON AI server module will reduce the preparation time of project documentation by 50-80% due to the automatic generation of user stories, acceptance criteria and functional requirements. The quality and structure of documentation will become uniform for all projects, which will reduce dependence on the human factor and reduce the number of errors during the development phase. The system will ensure quick updating of documents when requirements change, which will reduce technical debt and accelerate the launch of projects. Analysts and managers will be able to focus on more complex tasks, and development teams will receive clearer, standardized, and easily scalable artifacts. All this will increase the efficiency of development and allow the company to carry out more projects without increasing staff.
Haplusse Veronica
Purpose and description of task (project)
To develop a server core of the ZEON AI system that provides a full cycle of automatic generation of project documentation (user stories, acceptance criteria, functional requirements) based on input text descriptions and project templates. The backend must accept input data through the API, normalize and structure it, create a context for generation, orchestrate queries to AI models, perform postprocessing (splitting into entities, validating the structure, binding to modules/screens of the system) and save the results in a database with versioning. It is necessary to implement: a REST/GraphQL API for internal and external clients, a model of projects and artifacts (histories, criteria, requirements), queues for background generation and re-generation, a template system for different types of projects, access rights mechanisms (roles: analyst, PM, customer), audit of changes and logging of requests to AI. The backend should be scalable and fault-tolerant, support multi-tenant mode, ensure data security and the possibility of subsequent integration with Jira, Trello, ClickUp and other development management systems.
Note
Telegram @veronika_vgg