NebiusDigitalTwin

• 

  Digital Twin Security Center: Main dashboard overview and threat simulation entry points

• 

  Threat Simulation & Asset Inventory: Registering infrastructure nodes and executing AI threat analysis

• 

  Interactive Network Topology: Visualizing live asset connectivity and critical risk paths using vis-network

• 

  Nemotron AI Security Report: Real-time threat analysis and risk assessment powered by Nebius Token Factory and NVIDIA Nemotron.

• 

  AI Resiliency & High Availability: Automated failover mechanism active, seamlessly switching to Groq AI during assessmen

💡 Inspiration

Modern enterprise infrastructures face complex, multi-layered cyber threats where static vulnerability scanners fall short. I wanted to bridge this gap by building an intelligent, real-time Digital Twin Security Simulation Platform that models critical network topologies, simulates natural language threat vectors, and leverages advanced AI architectures to evaluate risks dynamically.

⚙️ What it does

NebiusDigitalTwin is an enterprise-grade security center built with a modular structure that allows security teams to:

• Manage an Asset Inventory Digital Twin: Register infrastructure nodes (Web Servers, Database Servers, Firewalls, Wireless Access Points) with custom CVE vulnerabilities and network segmentation properties.
• Simulate AI Threat Vectors: Input natural language attack scenarios (such as lateral movement or ransomware propagation) to receive global risk levels, precise attack paths, and prioritized mitigations.
• Orchestrate Multi-Provider AI Resiliency: Ensure zero downtime during critical security assessments via an automated fallback mechanism switching seamlessly between high-performance cloud models.

🛠️ How I built it

• Backend Architecture: Built using .NET 8 and strict Clean Architecture principles separated into modular layers (Domain, Application with MediatR CQRS pattern, Infrastructure, API, and Web).
• Data Persistence: Uses MongoDB for flexible document modeling of network nodes and simulation histories.
• AI Resiliency Engine: Powered by Nebius Token Factory (NVIDIA Nemotron) as the primary high-performance LLM provider, with Groq configured as an automated high-availability fallback to guarantee continuous availability.
• Frontend & Visualization: Developed with ASP.NET Core Razor Pages / MVC, featuring interactive UI elements and dynamic network topology mapping.

🚧 Challenges I ran into

• Designing a robust multi-provider AI failover mechanism that handles token limits or network timeouts gracefully without disrupting user sessions.
• Structuring a strict Clean Architecture in .NET 8 to maintain clear separation of concerns between domain logic, CQRS handlers, and infrastructure persistence.

🏆 Accomplishments that I'm proud of

• Successfully implemented an automated AI fallback strategy that maintains continuous telemetry and switches dynamically between provider states.
• Delivered a fully functional, enterprise-grade security simulation platform complete with an interactive web frontend and persistent MongoDB storage.

📚 What I learned

The critical importance of robust architectural patterns (Clean Architecture, CQRS) and resilient multi-provider LLM integrations when building production-ready AI applications for enterprise security domains.

Built With

• .net-8
• asp.net-core
• c#
• clean
• groq
• mongodb
• nebius-token-factory
• nvidia-nemotron