Systems Security & AI Engineer | Aspiring Graduate Researcher Building at the intersection of threat detection, industrial control systems (ICS/OT) security, and financial risk intelligence.
I design and implement security-hardened systems, edge telemetry validation gateways, and deep learning pipelines. My work focuses on:
- Industrial Control Systems (ICS/OT) Security: Zero-trust broker enforcement, PLC simulation, and Stuxnet-proof policy gateways.
- Agentic AI & Threat Intelligence: Multi-agent orchestration (LangGraph) for automated risk quantification and attack classification.
- Deep Learning Forensics: Document forgery classification (CNNs + Explainable AI) and video anomaly tracking (YOLOv8 + LSTM Autoencoders).
| Project | Research & Engineering Focus | Core Stack |
|---|---|---|
| aegis-ics | Zero-trust MQTT gateway featuring Stuxnet-proof AI-policy enforcer, trust scoring engine, and ESP32 hardware loop simulation. | Python, Flask, MQTT/TLS, ESP32, C++ |
| aegis-bank | Multi-agent threat detection and financial risk blast radius quantification pipeline built with LangGraph and FastAPI. | LangGraph, FastAPI, React, PostgreSQL |
| bergg-tactical-recon | Urban recon HUD and Wi-Fi heatmapping framework integrating ESP32 sensor scanning with MapLibre GL. | FastAPI, React, MapLibre GL, Docker, ESP32 |
| digital-surveillance-system | Real-time surveillance tracking and anomaly detection using YOLOv8, DeepSORT, and an LSTM Autoencoder. | PyTorch, OpenCV, YOLOv8, FastAPI |
| image-forgery-detection | Forgery detection pipeline processing Error Level Analysis (ELA) through fine-tuned CNNs and Grad-CAM spatial explanations. | PyTorch (ResNet), Streamlit, OpenCV |
- Languages: Python (Advanced), C/C++ (ESP32/Embedded), Javascript/ES6, SQL
- AI & Computer Vision: PyTorch, YOLOv8, OpenCV, LangGraph (Multi-Agent), Hugging Face, Streamlit
- Security & Network: MQTT/TLS, Zero-Trust Architecture, Cryptographic Verification (HMAC, JWT), WebSockets
- Infrastructure & Tools: Docker, Docker Compose, PostgreSQL, SQLite, Git, GitHub Actions (CI/CD)
- LinkedIn: Anshul R
- Email: anshul.ec23@sahyadri.edu.in
- Portfolio: Deployed live demos linked inside individual project repositories.
“Complexity is the enemy of security.” — Bruce Schneier. I build systems to prove it.