ARR update:

Building Sentinel DNA: Rethinking AI-Powered Security Investigations

I'm building Sentinel DNA, an AI security investigation platform focused on a problem I believe deserves more attention: turning security alerts into evidence-backed, explainable investigations.

Security teams don't just need another dashboard or an AI chatbot that summarizes alerts. They need systems that can connect evidence, enrich indicators of compromise, reconstruct attack sequences, map activity to MITRE ATT&CK, and explain why an investigation reached a particular conclusion.

That's the direction I'm taking with Sentinel DNA.

My core principles are:

• Evidence before conclusions.
• Explainable AI reasoning, with transparent confidence.
• Traceable evidence and investigation provenance.
• Controlled automation with human accountability.
• Security workflows designed for real-world SOC operations.

I'm currently focused on engineering credibility, security hardening, and independent validation before making bigger claims about readiness.

I'm interested in connecting with AI founders, cybersecurity engineers, security researchers, and people building reliable AI systems for high-stakes environments.

One question for the community:

What's the biggest gap you see in today's AI-powered security operations tools: investigation quality, evidence traceability, analyst trust, or safe automation?

I'd love to hear perspectives from people working on these problems.