Skip to content
View akrishnash's full-sized avatar

Block or report akrishnash

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 is supported. This note will only be visible to you.
Report abuse

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

Report abuse
akrishnash/README.md

Anurag K. Sharma

Secure AI for Mission-Critical Systems

AI Engineer · Scientist @ ADRIN, Department of Space

I build secure AI systems, autonomous agents, and quantum-safe network infrastructure for environments where failure is not an option — air-gapped networks, defence-grade VPNs, and critical 5G systems. My work sits at the intersection of applied AI, network security, and post-quantum cryptography.

Portfolio Writing GitHub Email


🚀 Flagship Projects

Project What it is Stack
VISHWAAS Secure enterprise VPN with a WireGuard control plane, approval-gated joins & connections, audit logs, and an AI assistant WireGuard · FastAPI · React
SecureAI Agent AI agent for cybersecurity — ingests logs, PCAP & CVEs, returns threat explanations, attack paths & mitigations GPT · RAG · Tool Calling
AirGapGPT (coming soon) Offline, air-gapped enterprise AI: chat, document RAG, RBAC, and audit logging — runs with zero internet Ollama · ChromaDB · FastAPI
5G Sentinel (coming soon) AI-assisted 5G core analysis — upload PCAP, get UE registration & NGAP explanations with security insights Free5GC · LLM · PCAP
PQC WireGuard (WIP) Post-quantum VPN: ML-KEM integration into WireGuard with handshake benchmarks ML-KEM · WireGuard

🧭 What I Work On

  • Secure AI systems — building AI that can be trusted in regulated, classified, and offline environments
  • AI agents — tool-calling agents for security operations and threat analysis
  • Network security — WireGuard, VPN orchestration, and zero-trust access control
  • Post-quantum cryptography — migrating real protocols to ML-KEM and benchmarking the cost
  • Air-gapped / offline AI — running capable models with no network dependency
  • 5G security — analysing 5G core signalling for anomalies and attacks

🛠️ Tech

Python FastAPI WireGuard OpenAI Docker Linux

📊 GitHub Signals

GitHub stats Top languages

✍️ Writing

I write about secure AI, AI agents, and post-quantum networking.

Article Topic Read time
Building Secure AI Agents with Tool Calling Grounding, bounded loops, trust levels & audit trails for production AI agents 14 min
Replacing X25519 with ML-KEM in WireGuard Hybrid KEM approach for post-quantum VPN — architecture, tradeoffs & benchmarks 12 min

View all writing

📫 Contact

Popular repositories Loading

  1. every-programmer-should-know every-programmer-should-know Public

    Forked from mtdvio/every-programmer-should-know

    A collection of (mostly) technical things every software developer should know about

  2. exo exo Public

    Forked from exo-explore/exo

    Run your own AI cluster at home with everyday devices 📱💻 🖥️⌚

    Python

  3. wg-vishwaas wg-vishwaas Public

    Post-quantum VPN — ML-KEM integrated into WireGuard, with handshake benchmarks and security architecture

  4. wireguard-ui wireguard-ui Public

    A UI for wireguard connection

    Python

  5. wireguard-connect-simulator wireguard-connect-simulator Public

  6. akrishnash.github.io akrishnash.github.io Public

    HTML