Proven leader with over twenty years of experience in cybersecurity architectures. 🚀

• GitHub: github.com/ajsween
  
• LinkedIn: linkedin.com/in/aaronjsweeney
  
• X (Twitter): @ajsween
  

Experienced cybersecurity professional with expertise in cloud architectures, database security, AI/ML deployments, and infrastructure as code. Skilled in Microsoft Azure, Docker, and security simulations. 🔒

• Microsoft - Cybersecurity Architect/Engineer (Current) Focused on AI Security solutions, including Microsoft Defender for AI Services and Azure-based security architectures.

• Cisco Systems Inc. - Public Sector Consulting Systems Engineer (2016 - 2022) Recognized as Engineer of the Quarter (Q2 FY18) for outstanding contributions in public sector consulting.
  

• Public Sector Consulting Systems Engineer of the Quarter (Q2 FY18) - Cisco Systems Inc. 🎖️
• Presidential Service Badge 🇺🇸

• analyze-prompt-intent GitHub 🔒
  Advanced AI safety analyzer that detects jailbreak attempts, prompt injections, and harmful content using local LLM inference via Ollama. Features multi-turn conversation analysis, ensemble mode with multiple models, advanced deobfuscation (Base64, ROT13, hex, Unicode), and GPT-OSS-Safeguard safety classifier integration.
  Technologies: Python 3.12+, Poetry, Ollama, OpenAI API, Chain-of-Thought reasoning, Microsoft Prompt Shields taxonomy.
  Key Features: Real-time threat detection, conversation trajectory analysis, crescendo attack detection, 30+ obfuscation techniques, exit codes for CI/CD integration.
  

• aicodegraph GitHub 📊
  Comprehensive Python repository analyzer that extracts code structure, detects dead code, and provides AI-powered insights through a GraphQL database and interactive web visualizer. Uses AST parsing to analyze classes, functions, imports, and relationships, with LM Studio integration for intelligent code analysis.
  Technologies: Python 3.14+, FastAPI, GraphQL (Graphene), Poetry, GitPython, AST parsing, OpenAI-compatible APIs, D3.js visualization.
  Key Features: Dead code detection, MCP tools for AI agents, interactive GraphViz visualizer, comprehensive reporting with AI recommendations, 30+ automated tests.
  

• bitnet-b1-58-arm-docker GitHub 📦
  Dockerfile for running BitNet-b1.58-2B-4T on ARM architecture (e.g., MacBook Pro M4 Max with Rancher Desktop and QEMU). Achieves 20-30 tokens/s performance.
  Technologies: Docker, Python, CMake, LLVM, Hugging Face CLI.
  

• defender-db-simulated-attacks GitHub 🛡️
  Bicep templates to deploy Azure SQL Managed Instance and resources for database security simulations. Includes security configurations, dynamic IP detection, and monitoring.
  Technologies: Azure CLI, Bicep, Nix, jq, nmap, sqlcmd.
  

• Virtual Reality and the Future of Tactical Military Training LinkedIn
  Article discussing military simulator technologies since World War II.

• Army Cyberspace Talent Management LinkedIn
  Article on effective talent management for Army's cyberspace dominance.

• Artificial Intelligence as a Force Multiplier in U.S. Military Information Campaigns CSIAC
  The article discusses the historical evolution of military information operations and how the integration of artificial intelligence can enhance these operations by improving information warfare capabilities, analyzing data, and generating content, while also addressing ethical implications.

Last updated: July 10, 2025 | Edit on GitHub ✏️aa