Quality assurance for an AI-driven platform

The platform was designed to transform how individuals consume and interact with digital content. By leveraging AI, this platform allows users to explore trending topics, generate personalized news summaries, and engage with content using text, voice, or image input.

Before our team joined the project, there was a lack of coordinated communication within the QA team, combined with limited access to resources like Figma, which created significant barriers to effective feature validation.

Challenges

• No test documentation was available on the project.
• Insufficient implementation of design according to Figma mockups.
• Need for testing AI functionality: verifying that the AI returns relevant responses.
• Different input methods, such as text input, voice input, and image recognition, need to be tested as ways of interacting with artificial intelligence.
• Insufficient QA team communication.

Solutions

• Created test documentation ( including detailed checklists and test cases). Implemented a new bug report format.
• Established close collaboration with the designer, leading to restructured mockups that enabled precise verification of new feature implementation.
• Conducted UI testing of AI-integrated product components, focusing on response relevance and correct processing of various input types.
• All input-related test cases were designed to include every relevant input method.
• Suggested daily stand-up meetings for better ticket distribution and team alignment.

• Chrome DevTools: Provides real-time inspection and debugging capabilities for web applications. Key features include element inspection, console logging, network monitoring, and performance analysis.
• Jira: A project management and issue tracking tool that facilitates efficient task management and team collaboration.
• Postman: Used for API testing, verifying responses, and validating data structures.
• LambdaTest: An advanced cloud-based solution for cross-browser testing, enabling teams to validate web applications across various browsers and operating systems.

• Incorporation of artificial intelligence for news generation and handling multiple input channels (text, voice, image).
• Ensuring cross-browser compatibility and responsive design for a consistent user experience across various devices.

• Up to 300 bugs were reported and fixed, covering issues across functionality, layout, and AI response accuracy.
• Up to 250 manual test cases were written, covering key user scenarios and edge cases.
• Relevance of AI responses across different input types (text, voice, image) improved by 17%.
• A new defect reporting format was introduced, resulting in up to 15% improvement in feedback turnaround time.