Traditional SAST scanners (Static Application Security Testing) lack deep code context, flooding teams with false positives that developers ignore. On the flip side, native LLM scanning is expensive, misses cross-file vulnerabilities, and fails to reach developers where they work.

Closing these gaps requires an intelligent engineering harness around the model. In this webcast, we will cover:

  • Building an AI Security Harness: How to gather cross-file context, self-verify findings, and deliver automated fixes directly to developer PRs

  • Eliminating False Positive Noise: Why combining AI context with traditional SAST cuts triage fatigue and catches vulnerabilities pre-production

  • The Enterprise Adoption Playbook: How to roll out incremental scanning, control model costs, and scale AI security across application code and IaC

You'll leave this session with a clear framework for building or scaling AI-assisted code security that developers will actually use.

Attendees are eligible to earn 1 CPE credit.

Generously supported by:

supported by logo
Speakers
speaker photo
Daniel Blazquez
Product Marketing for Code Security, Datadog

Daniel is a product marketing manager for Code Security at Datadog. Before moving into marketing, he was the product manager behind Datadog's first vulnerability detection product, taking it from concept to launch. Earlier in his career, he led marketing at Hdiv Security, a runtime application security company. The combination gives him an unusually hands-on view of how security tooling gets built, positioned, and adopted by engineering teams.

 

speaker photo
Joshua Delgado
Software Engineer, Datadog

Joshua is a software engineer on the Code Security product at Datadog. He works on application security and has contributed to the Static Analysis and Secrets products during his time at Datadog. Most recently he was involved in the GA of Datadog's AI Native Static Analysis product.