Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings
@ayinedjimi
ayinedjimi
Follow

Ayi NEDJIMI Consultants ayinedjimi

Cybersecurity Expert & AI Engineer | OSCP | Active Directory | CUDA/GPU Computing | Building AI-powered security tools

Block or report ayinedjimi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ayinedjimi /README.md

Ayi NEDJIMI

Cybersecurity Expert & AI Engineer | OSCP Certified

Sécuriser les infrastructures critiques. Construire l'IA de demain.

Website LinkedIn HuggingFace GitHub followers

Cybersecurity OSCP AI Engineer C++23 CUDA


About Me

name: Ayi NEDJIMI
role: Senior Cybersecurity Consultant & AI Engineer
certification: OSCP (Offensive Security Certified Professional)
location: France
company: Ayi NEDJIMI Consultants
website: https://ayinedjimi-consultants.fr

Expert in offensive and defensive cybersecurity with deep specialization in Active Directory security, AI/ML applied to security, and high-performance C++/CUDA development. I build tools that sit at the intersection of cybersecurity, artificial intelligence, and systems programming.

  • 20+ years securing critical infrastructures, 50+ audits completed
  • OSCP certified — hands-on offensive security expertise
  • Specialized in Active Directory, Microsoft 365, Azure, AWS & Kubernetes
  • Building AI-powered security tools and fine-tuning domain-specific LLMs
  • Publishing 100+ technical articles on cybersecurity and AI

GitHub Stats


Open Source Arsenal

A growing collection of 100+ security and AI tools I develop and maintain. Organized by domain, each tool is built to solve real-world problems I encounter during audits, research, and production deployments.


🛡️ AI Cybersecurity Tools

AI-powered tools that augment security operations — from threat intelligence to automated compliance.

Tool Description
ThreatIntel-GPT AI-powered threat intelligence analysis and correlation
LogParser-AI Intelligent log analysis with ML-driven anomaly detection
ComplianceBot Automated compliance checking against ISO 27001, RGPD, NIS2
VulnScanner-LLM LLM-powered vulnerability scanning and remediation advice
PhishingDetector-AI ML-based phishing email and URL detection engine
PolicyGenerator-AI AI-driven security policy generation from compliance frameworks
IncidentSummarizer Automated incident report summarization for SOC teams
CVE-Explorer-AI AI-assisted CVE exploration, impact analysis and prioritization
SOC-Assistant AI assistant for SOC analysts — triage, enrichment, response
SecureCodeReview-AI AI-powered secure code review for OWASP Top 10 vulnerabilities

🔒 Active Directory Security Audit

Purpose-built tools for deep auditing and attack detection in Active Directory environments.

Tool Description
LDAPRecon-AI AI-powered LDAP enumeration and security audit
ACLAudit-AI Active Directory ACL analysis — dangerous permissions detection
KerberosAudit-AI Kerberos configuration audit and weakness identification
GoldenTicket-Detector Golden Ticket and Silver Ticket attack detection
LateralMovement-Detector Lateral movement pattern detection across the domain
RemoteExec-Auditor Remote execution audit — PsExec, WMI, WinRM, DCOM
PrivEscAudit-AD Privilege escalation path audit in Active Directory
DelegationAudit-AD Kerberos delegation audit — unconstrained, constrained, RBCD
DCSyncAudit-AD DCSync rights audit and replication permission analysis
CredentialAudit-AD Credential hygiene audit — password age, reuse, service accounts

🚀 CUDA/GPU & MLOps Tools

High-performance GPU-accelerated tools and MLOps infrastructure for production AI pipelines.

Tool Description
CUDAEmbeddings GPU-accelerated embedding server with batched inference
GPUQuantizer Model quantization with CUDA — GPTQ, AWQ, GGUF
VRAMSwapper Intelligent VRAM/RAM offloading with multi-stream CUDA transfers
ADBloodHound-AI Active Directory graph analysis with AI-driven path finding
YaraGen-AI AI-powered YARA rule generation from malware samples
KQLHunter KQL query generation for Microsoft Sentinel threat hunting
ModelBench LLM benchmarking suite — latency, throughput, accuracy
DatasetForge Automated dataset creation and augmentation pipeline
HashCracker-GPU GPU-accelerated hash analysis and password audit tool
PacketSniffer-AI ML-driven network traffic analysis and threat detection

⚡ C++ Security Tools — 71+ Native Tools

A comprehensive suite of high-performance C++ security tools covering the full audit spectrum:

Active Directory Protocols

  • Kerberos audit & exploitation
  • NTLM relay detection
  • SMB security analysis
  • LDAP enumeration & audit
  • RDP security testing
  • DNS poisoning detection

Windows Security

  • Privilege escalation audit
  • Token manipulation detection
  • Registry analysis
  • Service permission audit
  • DLL injection detection
  • ETW trace analysis

Network Protocols

  • TCP/UDP packet analysis
  • TLS/SSL audit
  • ARP spoofing detection
  • DHCP security analysis
  • WMI audit tools
  • DCOM security analysis

All tools are built in modern C++ (C++17/C++20/C++23) with a focus on performance, safety, and minimal dependencies.


🧠 KVortex — Flagship Project

KVortex is a production-grade C++23 VRAM-to-RAM KV-Cache offloader for vLLM, enabling larger context windows and higher throughput on GPU-constrained hardware.

  • Multi-stream CUDA transfers — overlapped compute and data movement
  • Lock-free concurrent queues — zero-contention cache management
  • Adaptive eviction policies — LRU, LFU, and frequency-based strategies
  • vLLM integration — drop-in plugin for production inference servers
  • Benchmarked — measurable throughput gains on real-world LLM workloads
Architecture: CUDA C++23 | Lock-Free Queues | Multi-Stream Transfers | vLLM Plugin

🤗 HuggingFace — Models & Datasets

HuggingFace

4 Fine-Tuned Models · 50+ Datasets

Model Description
CyberSec-Assistant-3B 3B parameter model fine-tuned for cybersecurity Q&A, incident analysis, and threat assessment
RGPD-Expert-1.5B Specialized in GDPR/RGPD compliance — data protection questions and regulatory guidance
ISO27001-Expert-1.5B Fine-tuned for ISO 27001 implementation, audit preparation, and control mapping
M365-Expert Microsoft 365 security configuration, Conditional Access, and administration assistant

All models are trained on curated, domain-specific datasets built from my consulting experience and technical writing. Available on HuggingFace.


Tech Stack

Languages

C++ Python PowerShell CUDA Bash

AI & ML Frameworks

PyTorch vLLM HuggingFace LangChain

Security & Infrastructure

Kerberos Active Directory LDAP Kali Linux

Cloud & DevOps

Azure AWS Docker Kubernetes Git CMake


Latest Technical Articles

Cybersecurity & Pentest

AI & Machine Learning

Browse all 100+ articles at ayinedjimi-consultants.fr/tous-articles.html


Contact

🌐 Website ayinedjimi-consultants.fr
💼 LinkedIn linkedin.com/in/ayi-nedjimi
🤗 HuggingFace huggingface.co/AYI-NEDJIMI
🐦 X / Twitter @AyiNEDJIMI

Profile Views

© 2026 Ayi NEDJIMI — All rights reserved

Building security tools by day. Training models by night.

Popular repositories Loading

  1. ADReplicationInspector ADReplicationInspector Public

    C++ Active Directory replication health and security inspector

    C++ 1

  2. AzureArcAgentChecker AzureArcAgentChecker Public

    C++ Azure Arc agent configuration and compliance checker

    C++ 1

  3. BamDamForensics BamDamForensics Public

    C++ Windows BAM/DAM forensic analysis tool for execution evidence

    C++ 1

  4. PacketSniffer-AI PacketSniffer-AI Public

    AI-Powered Network Packet Analyzer with ML Anomaly Detection - Scapy/scikit-learn, detects port scans, DNS tunneling, DDoS, C2 beaconing, data exfiltration. For authorized security monitoring only.

    Python 1

  5. DCSyncAudit-AD DCSyncAudit-AD Public

    Active Directory DCSync Rights Auditor - Defensive security audit tool for identifying accounts with replication rights. Blue Team tool. For authorized security auditing only.

    Python 1 1

  6. ADauditor ADauditor Public

    Active Directory security audit toolkit for domain assessment

    PowerShell

AltStyle によって変換されたページ (->オリジナル) /