How AI Agents Transformed the QA Workflow
How the evolution of our internal AI eliminated repetitive manual tasks and standardized test case generation.
Case Overview
The Quality Assurance (QA) team is responsible for ensuring that applications perform flawlessly before reaching the end user. This involves analyzing requirements, aligning expectations with developers, validating user flows, and documenting every check performed throughout the lifecycle.
At Venturus, this workflow follows a structured sequence of essential steps to guarantee software excellence:
The Core Problem
While the process was well-defined, the final documentation stage remained a significant operational bottleneck.
Inputting data into the test management platform was a purely manual process:
- Every single check had to be registered individually, including titles, user actions, and expected system behavior.
- Information was entered cell-by-cell, consuming hours of labor and increasing the risk of human error.
As a result, the team spent more time documenting than analyzing, creating friction at a critical stage of the development cycle.
- Step-by-Step Copying
- Exhaustive Repetition
Defining the Challenge
To evolve the process, simple text automation wasn't enough. The central challenge was interoperability: ensuring that AI-generated content was fully compatible with our management platform, Microsoft Azure DevOps.
Key Challenges:
Format Compliance
Azure DevOps requires .csv files with rigid structural and syntax rules.
Contextual Adaptation
The solution needed to absorb the nuances and refinements of each specific project while maintaining the system’s required standards.
The Solution
To overcome these hurdles, the Venturus QA team developed an intelligent agent powered by ChapaVNT, our proprietary internal AI.
In practice, the solution integrates directly into the team's actual workflow at key stages:
- 1.Mind Map Generation: Receives the feature context and automatically organizes the verification flows and scenarios.
- 2.Context Refinement: The team adjusts and complements the information as needed, without losing standardization.
- 3.Standardized Documentation: Transforms the content into a consistent structure, ready for both human reading and management systems.
- 4.Azure DevOps Integration: Automatically generates the .csv file in the required format, enabling direct import, with no manual input.
The Results
75% reduction in planning time
Elimination of manual, repetitive tasks
Immediate standardization and Azure compatibility
Team focus shifted to strategy and high-level quality
Depoimentos
Lívia Fagali de Barros
"The AI agent brought significant time savings and standardization to our QA processes. We can now quickly generate inputs for the mind map presented in the QA Box, and create standardized test cases from that information in an Azure-ready format. Tasks that used to take days to complete are now done in just a few hours."
Emerson Ricardo Scarton
"Since I started working as a Software QA, I’ve always viewed test scenario planning as one of the most time-consuming and laborious tasks. With the agent, this process has been highly optimized. What used to take about 4 hours can now be done in approximately 1 hour, with a lower risk of human error and broader test scenario coverage."