Aligned with FDA CSA guidance for production and quality management system software — right-size your assurance activities based on process risk, not tradition.
Valkit.ai's proprietary CSAi (Computer Software Assurance Intelligence) framework activates when a GAMP category is selected during package creation. It optimizes test script creation for computer systems and software, aligned with FDA Computer Software Assurance guidance, and focuses on software-specific validation requirements.
The FDA recommends that manufacturers apply principles of risk-based testing where the selection, prioritization, and use of testing activities are based on analyzed risk. For high process risk features, this may mean scripted testing or a hybrid approach. For features that are not high process risk, unscripted methods such as scenario testing, error-guessing, or exploratory testing may be appropriate.
The LBA framework activates when no GAMP category is selected and enables risk-based test script creation for equipment, lab instruments, and processes. It implements the least burdensome testing approach described in FDA guidance, balancing validation rigor with resource optimization so that the burden of validation is no more than necessary to address the risk.
The Agentic AI framework processes unstructured data into structured requirements, converting documentation into contextual specifications. It handles up to 50,000 characters of input data and provides intelligent content parsing and organization. All AI features maintain human-in-the-loop controls — all workflows can be completed without AI use or intervention.
AI-optimized test creation aligned with FDA CSA guidance
Least burdensome assurance for equipment, instruments, processes
Unstructured data to structured requirements (up to 50,000 chars)
Systematic identification of foreseeable software failures
Flexible testing methods scaled to risk level
All AI outputs require human review; all workflows work without AI
The FDA CSA guidance (February 3, 2026) describes computer software assurance as a risk-based approach for establishing confidence that software is fit for its intended use. Valkit.ai's CSAi and LBA frameworks operationalize this approach within the platform.