B.Tech CSE @ VIT Chennai (2023–2027) · CEH Certified · Cybersecurity Essentials (Cisco)
I build systems that sit at the intersection of security, machine learning, and infrastructure resilience. My work spans from kernel-level eBPF instrumentation to adversarial ML defense to production healthcare platforms — always with the same question: "How does this break, and how do I prevent it?"
Currently interested in: Application Security · Adversarial ML · Kernel Security · Cloud-Native Resilience · Zero-Trust Architecture
Languages
AI / ML & Security
Frameworks & Infrastructure
|
Autonomous eBPF resilience engine for Kubernetes. Predicts cascading failures using kernel-level probes and Monte Carlo simulations before outages occur.
|
Pre-inference adversarial defense for medical imaging AI. Detects & neutralizes FGSM, PGD-40, DeepFool attacks on chest X-ray models.
|
|
Enterprise healthcare platform with zero double-booking concurrency (PostgreSQL row-locking), 9 Gemini AI clinical pipelines, and doctor accountability.
|
🔏 MetisZero-trust PII redaction engine — dual-layer detection with fine-tuned RoBERTa NER (96.6% F1) + 11 regex patterns. AES-256 reversible vault.
|
|
Real-time fraud detection with hybrid ML risk scoring + NVIDIA NIM LLM (Llama-3.1-70b) explainability for compliance teams.
|
Modular network reconnaissance engine — TCP SYN scanner, OS fingerprinting, TLS analysis, banner grabbing, and async CVE correlation.
|
🔴 Certified Ethical Hacker (CEH) — EC-Council | 🔵 Cybersecurity Essentials — Cisco
"Build systems that are hard to break. Then try to break them anyway."