What I Do
From building detection rules to deploying full SOC pipelines, I focus on practical, hands-on security engineering.
- ●Deployed end-to-end SOC pipelines - From log ingestion to alert generation
- ●Performed alert triage and incident investigation
- ●Correlated multi-source logs to reconstruct attack chains and identify detection gaps
- ●Developed KQL-based detection rules for brute-force, privilege escalation, and reconnaissance
- ●Tune and validate signals to reduce noise and improve reliability
- ●Mapped detections to MITRE ATT&CK for structured coverage analysis
- ●Design systems with controlled access and clearly defined security boundaries
- ●Validate how systems behave under realistic usage and threat conditions
- ●Documented deployment architectures for production use cases
- ●Published research on AI-powered deepfake voice detection
- ●Built Python automation pipelines for security workflows
- ●Develop internal tools and workflows to improve visibility and analysis
Featured Work
Two projects that represent my best work — one in product engineering, one in research.
Problem
Pentest outputs lack structure — events are disconnected, making it hard to trace how an attack actually progressed and where it led.
What was built
An attack chain intelligence platform that transforms flat pentest output into structured, interactive, MITRE-mapped attack chain views.
Key outcomes
- ●Combines AI-driven interpretation with deterministic flow and validation logic
- ●Reconstructs attack sequence, context, and privilege evolution across events
- ●Bridges fragmented outputs into a unified, dependency-aware attack understanding
- ●Flags inconsistencies, invalid transitions, and incomplete attack paths
- ●Designed as a scalable system for real-time ingestion and analysis
Problem
AI-generated voices make it increasingly difficult to trust voice-based interactions, enabling high-impact vishing and impersonation attacks.
What was built
A real-time voice analysis system designed to identify synthetic and manipulated speech in live and recorded scenarios.
Key outcomes
- ●Converts audio signals into frequency-domain representations for pattern analysis
- ●Trained ML models (CNN/LSTM) on extracted audio features to distinguish real vs synthetic speech
- ●Evaluates temporal and spectral characteristics to detect inconsistencies in generated voices
- ●Research-backed system with focus on practical detection of voice-based attacks
Experience
My professional journey in security operations and detection engineering.
SOC Analyst
— CyArt Tech LLPJan 2026 – PresentCurrent
- ●Deployed and operated SOC workflows across detection, investigation, and response
- ●Build and refine detection logic based on real attack patterns and observed behavior
- ●Handle end-to-end attack scenarios — from simulation to analysis and containment
- ●Contributing to development of an attack intelligence system (Chain Discovery)
Infrastructure Security
— FibmeshAug 2025 – Dec 2025
- ●Designed and validated secure access architectures for remote connectivity use cases
- ●Built production-like deployments with controlled exposure and access restrictions
- ●Documented deployment guides and recorded demo videos
Cyber Security Associate
— InfoSoft ComputersMay 2024 – Present
- ●Design and structure learning modules focused on practical security concepts and tool usage
- ●Support participants in setting up environments, troubleshooting and understanding real-world scenarios
- ●Contribute to cybersecurity training programs through workshop delivery and hands-on lab support
B.E. Information Technology (Honors in Cyber Security & Privacy)
Savitribai Phule Pune University | 2021 – 2025 | CGPA: 7.83