Ayi NEDJIMI - Open Source Security Arsenal
**Cybersecurity Expert & AI Engineer OSCP Certified 20+ years experience** 100+ open-source tools for cybersecurity, Active Directory auditing, AI/ML, and GPU computing. All tools are freely available on GitHub.
Quick Navigation
| Category | Tools | Language |
|---|---|---|
| AI Cybersecurity | 10 tools | Python |
| AD Security Audit | 10 tools | Python |
| CUDA/GPU & MLOps | 10 tools | Python/C++ |
| C++ Security Tools | 71+ tools | C++ |
| KVortex | Flagship | C++23 |
| HuggingFace | 4 models, 50+ datasets | - |
Why These Tools?
Most enterprise security tools are expensive and proprietary. This arsenal provides free, open-source alternatives for:
- Red & Blue Teams needing fast Windows forensics and AD auditing
- SOC Analysts requiring AI-assisted triage and log analysis
- AI/ML Engineers working with large language models on GPU infrastructure
- Compliance Auditors assessing ISO 27001, ANSSI, and GDPR requirements
Featured Tools
ThreatIntel-GPT
AI-powered threat intelligence analysis using large language models. Processes threat feeds, generates IOC reports, and correlates indicators across STIX/TAXII, MISP, and OpenCTI sources. Supports MITRE ATT&CK mapping and automated PDF/HTML reporting.
pip install threatintel-gpt
threatintel-gpt analyze --source feeds.json --output report.html
| View on GitHub | Download v1.0.0 |
KVortex
Production-grade C++23 VRAM to RAM KV-Cache Offloader for vLLM. Multi-stream CUDA transfers achieving 22.4 GB/s bandwidth. Enables larger context windows without additional GPU hardware.
auto engine = KVortexEngine::create(config);
engine->save_blocks(tensor, block_ids);
engine->load_blocks(block_ids, output_tensor);
| View on GitHub | Download v1.0 |
CUDAEmbeddings
GPU-accelerated embedding server with CUDA optimization for real-time vector generation. Processes 10,000+ embeddings/second on consumer-grade NVIDIA GPUs.
pip install cuda-embeddings
cuda-embeddings serve --model all-MiniLM-L6-v2 --device cuda:0
| View on GitHub | Download v1.0.0 |
About
I am Ayi NEDJIMI, a cybersecurity consultant and AI engineer based in France. With OSCP certification and 20+ years of experience, I specialize in:
- Offensive & Defensive Security: Active Directory, Kerberos, NTLM, Windows internals
- AI for Security: LLM-powered threat analysis, automated audit tools
- GPU Computing: CUDA optimization, model quantization, VRAM management
- Compliance: ISO 27001, RGPD/GDPR, ANSSI, security policy automation
Frequently Asked Questions
Are all tools free? Yes, all tools are open-source under permissive licenses.
What OS is required? Most C++ tools target Windows 10/11 and Windows Server 2019+. Python tools are cross-platform.
Do I need a GPU? CUDA tools require an NVIDIA GPU (GTX 1080 minimum). All other tools run on CPU.
Can I use these in production? Yes. Tools marked v1.0+ are production-tested in real enterprise environments.
Contact
- Website: ayinedjimi-consultants.fr
- GitHub: github.com/ayinedjimi
- LinkedIn: linkedin.com/in/ayi-nedjimi
- HuggingFace: huggingface.co/AYI-NEDJIMI
2026 Ayi NEDJIMI - All rights reserved