As VSCO’s development organization grew, its QA process became a bottleneck. Testing often occurred late in the sprint and was largely limited to basic sanity checks. Time constraints frequently narrowed regression coverage to the feature under active development, increasing the risk that related functionality would go untested.

QA engineers were not consistently embedded within development teams, limiting their exposure to VSCO’s product direction, customer needs, and the intent behind new features. Test planning depended heavily on documented requirements, making it harder to identify missing scenarios while those requirements were still evolving.

The increasing complexity of the product and test suite created another challenge. Evaluating changes only through visible application behavior provided an incomplete view of potential impact, while running every test case within each sprint was no longer practical. VSCO needed earlier feedback, stronger coverage, and a consistent way to prioritize testing without slowing releases.

Krasamo QA engineers work directly within VSCO’s development teams, collaborating with developers throughout each sprint. This close integration gives testers a stronger understanding of the product, its users, and the goals behind each feature.

This knowledge allows the team to go beyond validating written requirements. QA engineers can identify gaps in functionality, anticipate how customers might use a feature, and begin testing when working code becomes available, even while some requirements are still being refined.

Krasamo engineers also have access to VSCO’s source code repositories and development tools. Because many team members have software development experience in addition to testing expertise, they can examine the code affected by a change and build coverage around both the underlying implementation and user-facing behavior.

Testing priorities are determined through change-impact and risk analysis. Depending on the change, the team uses smoke tests, sanity checks, targeted regression testing, and white-box testing where appropriate.

The QA team also uses VSCO’s internal AI agents and models to identify, draft, and review test cases based on source code, requirements, and the current mobile application. This supports faster verification and gives QA engineers more time for testing that depends on deeper product knowledge.

Testing now begins when working code becomes available instead of waiting for requirements and features to be finalized. Embedded QA engineers use their knowledge of VSCO’s product and customers to identify missing scenarios and functional gaps earlier in development.

Access to source code gives the team visibility into the underlying changes. This supports a broader, change-specific testing approach rather than relying mainly on basic sanity checks or executing the entire test suite for every change. Regression work can focus on the workflows and components most likely to be affected.

VSCO’s AI agents and models further support test-case identification, drafting, and review. This gives QA engineers more time to evaluate the product-specific scenarios and nuanced behavior involved in image and video manipulation.

Together, risk-based test selection and AI-assisted test design help keep QA effort from growing in direct proportion to the size and complexity of the test suite.