AMBER BENNOUI
Amber Bennoui

AI · Cybersecurity · Product · Strategy

Amber Bennoui

AI and cybersecurity product leader with 15+ years of experience across companies including F5 Networks, Datadog, Chainguard and DataRobot. Co-founded the AI Security Alliance (AISECA), and conducts ongoing research into AI supply chain security. Offers product management coaching, consulting, fractional CPO services, and AI and security strategy guidance.

Boston, MA · Speaking, coaching, consulting & fractional CPO inquiries welcome

Socials

Speaking

Agents in the AM

DataRobot — recurring hands-on workshop

A live lab on building and shipping AI agents. Delivered five times across London, Washington DC, Boston, Dallas, and New York, with more sessions planned.

The Hard Part of AI Agents No One Talks About

DataRobot

Why production-ready agents come down to developer experience, agent identity, observability, and token economics, not the model or the prompt.

Beyond Guardrails: Building Secure-by-Design AI Culture

Boston Security Meetup, hosted at 7AI

Three pillars for teams shipping AI: secure-by-default design, agent identity as a first-class citizen, and supply chain instrumentation across the full lifecycle.

Intuit Open Source Platform Engineering & AI Meetup

Mountain View, CA

On the gap between an agent demo and something a company can actually run in production. Shared the stage with speakers from Cursor and RX-M.

DataRobot × NVIDIA — Federal Session

Washington, DC

Enterprise agent deployment for the public sector.

Boston Tech Week

Boston, MA

Ran a hands-on DataRobot agent-building workshop (standing room only, 30+ agents built in under two hours), spoke on storytelling and sacrifice in product development with fellow Simmons University alumnae, and moderated a security panel and mixer with Scrut Automation, DigitalOcean, Citizens, and Monad.

Guardrails for Agents: Security Patterns for Platform Teams Running AI

KubeCrash Spring 2026 (virtual)

Panel moderated by Beatriz Datangel Rodgers, alongside Lisa Tagliaferri and fellow practitioners.

Tackling Tool Sprawl in Cybersecurity: Strategies for Streamlined Defense

On consolidating security tooling without losing coverage.

Building an Inclusive Culture with ERGs

DASH — Datadog's annual conference

Panel with three fellow Datadog ERG leaders on the programs behind the impact.

Security Observability Day

Datadog

Product demo: "Uncover Critical Vulnerabilities Worth Fixing."

Podcasts

AI Security Is Just as Vague as "Cloud Security," But With Sparkle Emojis

Bare Knuckles and Brass Tacks

Recorded just after RSA Conference. On why so much of what's sold as AI security is really cloud security with new branding, and the discoverability problem underneath it.

Iron Mind Podcast

with host Joshua Brumley

Data protection fundamentals for law firms: access controls, encryption, and verifying who's actually on the other end of a request.

Writing

DataRobot Blog

Regular contributor on agentic AI, developer tooling, and building for production.

One Year Later: Signing CISA's Secure by Design Pledge

Chainguard, with Matt Moore and Anushka Iyer

State of DevSecOps

Featured in Datadog's annual report

On why minimal, distroless images shrink the attack surface.

Research

The Devil Wears Markdown: Detecting Malicious Intent Hidden in AI Agent Skill Instructions through Hybrid Static, Dynamic, and Agentic Analysis

with Thanh M. Truong-Le, Hoang-Ly Nguyen, Thang Ho, Thai-Minh Truong, and Duc-Ly Vu — unpublished

Trust Nothing, Install Nothing Unscanned: A Practitioner's Framework for AI Artifact Security

with Ly Vu, Thang Ho, Thanh Truong, and Ly Nguyen — unpublished

Transcranial Ultrasound Detection of Intracranial Hemorrhages: Time-Frequency Analysis with Empirical and Variational Mode Decomposition

Journal of the Acoustical Society of America

Transcranial Ultrasound Hemorrhage Detector

Acoustical Society of America

Community & Leadership

AISECA — AI Security Alliance

Founding Board Member

Building a practitioner-led AI security maturity framework, vendor-agnostic by design.

MEACS — Middle Easterners & Africans in Cyber Security

Founding Member

Cyber Women of Boston

Organizer

Speaking Topics

  • Getting AI agents from pilot to production
  • Developer experience for AI agent builders
  • AI and product strategy
  • Enterprise AI ROI and observability
  • AI supply chain security & AI artifact risk
  • Secure-by-design culture for AI-first teams
  • Agent identity, authorization, and governance at scale
  • DevSecOps tool consolidation
  • Building inclusive teams in security and tech