Ayi NEDJIMI - Cybersecurity & AI Tools

100+ open-source cybersecurity, AI, and GPU computing tools

View on GitHub

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:


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:

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


2026 Ayi NEDJIMI - All rights reserved