SHADOWFIREWALL
Architecte de défense adaptative
Profil professionnel
ShadowFirewall crée des pare-feu comportementaux fondés sur des modèles de comportements hostiles. Avec une expérience en IA et en systèmes embarqués, il conçoit des défenses réactives qui apprennent du trafic.
Compétences clés
Pare-feu adaptatifs; modélisation des comportements hostiles; détection d’anomalies; micro-segmentation; sécurité OT/SCADA
Réalisations majeures
Conception du premier pare-feu comportemental d’une usine intelligente européenne.
Articles de SHADOWFIREWALL
When Shared Blocks Remember Too Much: Cloudflare’s Cross-Tenant Leak Warning
A remediated storage flaw in Cloudflare’s Containers stack shows how multi-tenant isolation can fail at the disk layer, even when compute boundaries stay intact.
When One Customer Could Read Another's Dust: Cloud Storage Isolation Fails at the Block Layer
A cross-tenant exposure in Cloudflare's container stack shows how a shared disk optimization can turn tenant isolation into a data-remanence problem.
AI-Assisted Commits Are Turning Secret Leakage Into an Identity Crisis
GitGuardian’s 2026 State of Secrets Sprawl Report points to a troubling pattern: AI-assisted commits are leaking secrets at roughly twice the rate of human-written ones, shifting the fight from code quality to credential control.
Keeper’s AI Warning Lands Where Security Actually Breaks: Identities and Secrets
A Keeper Security commentary puts the spotlight on a familiar weak point in modern environments: who can authenticate, what they can reach, and which secrets they carry.
Forgotten Microsoft 365 Accounts Became the Soft Target in a Cloud-Scale Identity Push
A TeamFiltration-linked campaign tracked as UNK_CondorFiltration shows how weak account hygiene, not flashy malware, can still turn Microsoft 365 into an access point.
When an Email Alias Becomes a Trust Token in GitLab
A private incoming-email address can act like a hidden authorization path for work-item creation, and in permissive GitLab setups that trust boundary may ripple into branch governance and CI/CD control.
One Kubernetes Manifest, One Big Cloud Risk: The Config Connector Trap
A Google Cloud integration designed to simplify infrastructure can become a privilege amplifier if its controller is trusted more than the person submitting the manifest.
When Login Risk Becomes the Gatekeeper, Identity Security Splits in Two
Adaptive authentication is no longer one market with one purpose - workforce access control and fraud detection now operate as different control planes with different signals, tradeoffs, and failures.
The Hidden Identity Layer Security Teams Are Still Buying in Pieces
A 2026 roundup of machine-identity tools points to a deeper problem: enterprises are still splitting certificate control, cloud automation, and workload authentication into separate jobs.
The Quiet Arms Race Behind Just-in-Time Access
A 2026 roundup of JIT tools points to a bigger security shift: organizations are trying to shrink privileged access into short-lived, auditable windows instead of leaving it always on.
The Real Breach Was the Consent Screen: Google Workspace’s Quiet Attack Surface
A webinar on Google Workspace incidents points to a familiar but dangerous pattern: social engineering, malicious OAuth apps, and a narrow window where fast containment can matter more than perfect hindsight.
AI Agents Get a Stricter Gatekeeper as Identity Control Moves Up the Stack
Aembit’s support for Okta Cross App Access points to a quieter shift in enterprise security: letting AI agents act, but only inside policy-bound identity paths.
The Quiet Power Grab Behind Cloud Directory Choices
A roundup of eight cloud directory services is really a map of where organizations want identity, access, and device control to live after Active Directory starts losing its central role.
Biometrics Are Splitting Into Four Battles, and Fraudsters Know It
Identity proofing, liveness checks, post-login monitoring, and voice anti-spoofing are emerging as separate control layers, not one all-purpose answer.
Microsoft Puts SMS Sign-In on the Clock in Entra ID
A planned 2027 retirement of SMS as a primary sign-in method is pushing Microsoft Entra ID tenants toward passkeys, FIDO2 keys, and other phishing-resistant login paths.
Microsoft Sets a 2027 Clock on SMS Sign-In, and Entra Admins Have Work to Do
Microsoft is steering Entra ID tenants away from text-message sign-in and toward passkeys, turning an old convenience feature into a migration deadline.
The Cookie That Could Tie AI Ads to a User's Browser Trail
A newly observed OpenAI advertising cookie shows how ordinary browser state can become a privacy-sensitive bridge between one site and another, even when identity linkage is not proven.
The Hidden Fight Behind MFA: Why Fit Beats Fame in Identity Security
A 2026 MFA roundup puts the spotlight on a bigger truth: the safest login control is the one that matches the environment, especially when phishing-resistant methods and legacy app coverage diverge.
Passkeys Are Not One Product - They Are a Security Choice About Control
A new wave of passwordless tools is less about replacing passwords and more about deciding whether trust lives on the device, in the identity provider, or inside a verified onboarding flow.
When a Photo Parser Becomes the Weakest Link in the Cloud
A reported image-decoding flaw class shows how HEIF, HEIC, and AVIF handling can widen risk across collaboration and code-hosting systems that trust uploaded media.