OptiQA

Project: OptiQA

Tagline: Intelligent, AI-powered quality assurance that keeps pace with modern development.

Team: Quantum

Project Status: In Development

Overview

Software quality assurance today is often manual, repetitive, and slow — creating a bottleneck in the software development lifecycle. QA teams spend countless hours writing test cases, running regression tests, and documenting results, while development continues to accelerate. OptiQA is an AI-powered platform that automates critical parts of the QA process. It generates, executes, and validates test cases intelligently, ensuring higher accuracy and faster release cycles. By combining NLP, ML models, and automation pipelines, OptiQA transforms QA from a manual chore into a continuous, intelligent process. Our vision: QA that’s adaptive, fast, and reliable — keeping up with the speed of innovation.

The Problem

Manual QA is time-consuming and does not scale with modern agile development. Test coverage is often incomplete, leaving gaps that result in production bugs. QA engineers spend too much time on repetitive tasks (regression, log reviews, bug reporting). Companies face higher costs and risks due to human error and inefficient QA cycles.

Our AI Solution

OptiQA is an AI-powered Quality Assurance automation platform designed to simplify and accelerate the testing process for web applications. Its purpose is to:

• Reduce manual testing effort.
• Automatically generate, execute, and validate test cases using AI.
• Enable QA engineers and developers to test faster, smarter, and more efficiently.
• Provide detailed, real-time test results and maintain prompt/test history for traceability.

Key Capabilities

• AI-driven Test Generation: Users can input natural language prompts; AI converts them into executable test cases.
• Automated Execution: Playwright executes the generated test scripts automatically.
• Result Visualization: Users can view detailed results for each test run — including pass/fail status, screenshots, and logs.
• Prompt History Management: All user prompts and corresponding results are stored and retrievable.
• User Authentication: Secure login/signup using JWT-based authentication.

Technology Stack

Workflow Description

1. User Registration/Login → User signs up or logs in.
2. Prompt Submission → User enters URL + natural language steps (e.g., “Login to Gmail and verify inbox count”).
3. AI Processing → Backend sends the prompt to GLM 4.6 model to generate Playwright test script.
4. Test Execution → The system executes the generated code using Playwright.
5. Result Storage → Logs, status, and screenshots are saved in the database.
6. Result Visualization → User views the detailed report and test history from the frontend.

System Architecture

Architecture Overview

OptiQA follows a modular microservice-inspired structure:

1. Frontend (Flutter)
   Handles user interaction, authentication, prompt submission, and data visualization.

2. Backend (Node.js + Express)
   Manages user requests, invokes GLM 4.6 for AI processing, executes Playwright tests, and interacts with PostgreSQL.

3. AI Layer (Hugging Face's GLM 4.6)
   Generates test scripts based on natural language prompts.

4. Database Layer (PostgreSQL)
   Stores users, prompts, test results, execution logs, and screenshots.

5. Playwright Test Engine
   Executes generated test scripts and sends results back to the backend for storage and visualization.

Architecture Diagram

High Level Design (HLD)

The system follows a modular, scalable architecture that connects frontend, backend, AI, and database layers.
The frontend communicates securely with backend APIs, which orchestrate test execution, AI analysis, and reporting.

HLD Diagram

LLD Diagram

Key Components

• User Management Service: Handles registration, authentication, and user profile management.
• Prompt Service: Accepts and stores user-submitted prompts; links each prompt to a user.
• AI Response Service: Interacts with an AI model to generate responses for submitted prompts.
• Test Result Service: Runs automated test cases on AI-generated outputs and stores results.
• Chat Logging Service: Records all user–AI interactions (both user and AI messages).

Database Design

Challenges & Learnings

Converting unstructured requirements into testable cases. Handling variability in logs and error patterns. Ensuring scalability across different projects and teams. Combining NLP + rule-based logic works better than pure ML in early stages. QA automation isn’t just about test execution — it’s about actionable insights. Building a polyglot system (Flutter + Node.js + Python) requires tight orchestration.

Future Roadmap

Short-term goal - Implement regression automation and intelligent log parsing. Medium-term goal - Fine-tune NLP models to generate end-to-end test suites. Long-term goal - Adaptive QA assistant that self-learns from previous test cycles and suggests new edge cases automatically.

Repository & Live Demo

GitHub Repository: FE - https://gitlab.enovate-it.com/enovate/hackathon/optiqa-fe BE - https://gitlab.enovate-it.com/enovate/hackathon/optiqa-be

Categories: #project #ai-hackathon-2025 #category-QA #category-automation #category-ai