David Weidman
Founder, Idea Security Company (SenTeGuard) • Policy &
Engineering
Harvard Kennedy School (Master in Public Policy, International & Global Affairs)
Find me on
Background
I'm an AI security founder and policy-oriented engineer focused on "idea security": how meaning moves through humans, models, and workflows, and how organizations can enforce boundaries when language models make it easy to transform and transmit knowledge.
I graduated from the United States Military Academy at West Point ('16) with dual degrees in Computer Science and Mathematical Sciences. Early on, I built machine-learning systems for strategy games and worked on applied ML research in communications systems.
In the U.S. Army, I served in operational and technical roles, including leading teams in cyber-relevant missions and advising senior commanders. That experience shaped how I think about real-world constraints: uncertainty, adversaries, and the need for systems that work under pressure.
I graduated from the Harvard Kennedy School Master in Public Policy program (International & Global Affairs). I'm building SenTeGuard, a company whose mission is to protect organizations' most sensitive ideas.
What I work on
I build security products and policy frameworks for the "meaning layer" of AI systems: preventing sensitive knowledge from leaking when employees summarize, paraphrase, translate, or repackage internal information through LLM tools.
My focus is practical: enterprise AI adoption that holds up under real compliance constraints (privacy, export controls, regulated R&D, and national-security environments), without slowing teams to a crawl.
Current focus
- SenTeGuard: A security layer for enterprise GenAI that detects, warns, flags and blocks risky content flows at the point of use, especially semantic leakage when employees summarize, paraphrase, translate, or abstract sensitive material.
- Moyo: An "information-space mapper" that crawls and maps an organization's public-facing knowledge surface, then measures how close internal concepts are to becoming externally reachable through LLMs, search, and retrieval tools. I use it for outside-in risk discovery and targeted AI red-teaming that feeds directly into concrete controls and SenTeGuard deployments.
- Joseki:Wrapperhub: A prompt-spec marketplace and version-control layer for reusable AI workflows. The goal is to make prompts and agentic "recipes" auditable, testable, and governable: with changelogs, licensing, security scoring, and organization-level guardrails for how prompts are shared and reused.
- Idea Security Verification Framework (ISVF): A governance and controls framework for meaning-based security and "boundary maintenance," designed to align with enterprise compliance constraints and export-controlled technical data regimes.
Skills
Technical
- Security engineering for GenAI systems (DLP, content risk, guardrails)
- Applied ML / LLM evaluation, red-teaming, and safety testing
- Backend systems, enterprise integrations, and deployment design
- Product strategy for B2B security and compliance tooling
Policy & leadership
- AI governance, national security, and risk frameworks
- Cross-functional leadership: technical teams and senior stakeholders
- Writing and briefing for technical + non-technical audiences
- International experience across defense and security contexts
Languages
In decreasing competence: English, Korean, French, Japanese, Italian, Mandarin
Contact
Email: [email protected]
LinkedIn: Personal • SenTeGuard
Signal: djwide.13
Based in Austin, TX • Open to speaking, advising, and enterprise pilots.