The preemptive observability engine is active during pre-production testing to catch AI-generated bugs and other coding issues. Digma has unveiled Preemptive Observability Analysis, an observability engine designed to reduce bugs introduced by AI code generation and other coding issues. The commercial version, introduced February 5, follows a limited-capability free version. Digma said its engine provides engineering teams with code-level insight into the root cause of software development problems while adding AI-driven suggestions to resolve performance issues, architectural flaws, and problematic runtime behaviors. Preemptive Observability Analysis (POA) engine is active during the pre-production testing stage of the software development lifecycle. In addition to tackling AI-generated bugs, Preemptive Observability Analysis engine also addresses bugs in human-generated code that cause performance issues and SLA (service level agreement) degradations, Digma said. The tool will be particularly transformative for organizations in highly transactional environments such as fintech, e-commerce, and retail, said the company. Many companies have relied on application performance monitoring tools to highlight service problems, SLA failures, and monitor production status, Digma said. These tools can send alerts when services slow down or fail in production, or when thresholds are breached, but they do not offer insight into the root cause of problems and are limited when used in non-production environments, according to Digma. By shifting the focus to finding issues earlier in the software development lifecycle with preemptive observability, teams can prevent problems rather than deal with their impact later, the company said. Preemptive observability also is intended to help enterprises reduce cloud costs through early identification of performance and scaling issues. The commercial release of Digma’s Preemptive Observability Analysis engine includes the following capabilities: Pattern-based issue identification before code reaches production. AI-driven fix suggestions based on runtime behavior analysis. Team collaboration insights to prevent code conflicts between teams. Cloud cost optimization through early detection of scaling issues. Management dashboards for non-coding engineering leaders. Sandbox environment for evaluation without deployment. Generative AISoftware DevelopmentApplication Security