Case Study: Semgrep reduces false positives and boosts code security with Claude

How Semgrep delivers AI-powered code security with Claude in Amazon Bedrock

Semgrep, a cybersecurity company, faced the significant industry challenge of security tools generating an overwhelming volume of false positive alerts, which wasted developers' time and resources. To address this, they turned to the vendor Claude and leveraged its capabilities through Claude in Amazon Bedrock to bring AI-powered analysis to their platform.

By implementing Claude, Semgrep achieved a 20% reduction in false positive security alerts out of the box, with a 92% user agree rate on the safe-to-ignore labeling. The solution from Claude also delivered a 16% higher accuracy in identifying false positives and 17% better performance in component tagging compared to their previous model. This allows Semgrep to confidently process and analyze thousands of security findings for their customers daily.

Semgrep

Chushi Li

Product Marketing Manager