Cybersecurity × AI engineering student. Five internships across security operations, full-stack, and DevOps — each one ended with something deployed. I build AI & security systems that actually run.
01 // Identity
I'm Sabarish R — a third-year B.Tech CSBS student at Panimalar Engineering College, Chennai (class of 2028), working across cybersecurity, applied AI, and full-stack engineering.
Threat detection platforms, LLM guardrails, SOC automation, MERN products — I treat every build as something that has to survive contact with real traffic.
02 // The Record
03 // The Vault
Verifies skills from proof, not self-reports: evidence pipeline over GitHub/LeetCode/resume signals, a 211-skill prerequisite DAG, and a fine-tuned bi-encoder lifting Recall@5 from 56.9% to 93.8% as a 32MB on-device model.
Phishing/BEC detection with an adaptive ensemble (XGBoost + LightGBM + DistilBERT) trained on 30k+ emails, SPF/DKIM/DMARC header forensics, live threat map, and a URL-guard browser extension.
Guardrail layer for LLM apps and agents: 10 detectors, an allow/sanitize/quarantine/block verdict engine with Redis telemetry, MCP tool-call security, and a live SOC console.
Maps events to MITRE ATT&CK techniques, correlates IOCs, and ships monitoring rules with live ingest, alerting, and incident-response playbooks. 31-test suite, CI green.
Co-developed platform detecting malicious phishing sites, malware, and anomalies, with a client extension and centralized cloud telemetry.
04 // The Stack
05 // Credentials & Contests
06 // The Archives