I work at the intersection of adversarial AI and security engineering — red teaming production AI systems at JPMorganChase before they reach end users. Traditional security reviews don't cut it for AI. My job is to find the things that change deployment decisions.
What I'm working on:
- Leading adversarial assessments of enterprise GenAI use cases — prompt injection, jailbreaks, intent deviation, alignment failures
- Building and operationalizing red teaming toolkits and evaluation frameworks used across the AI Red Team
- Running adversarial and behavioral evaluations across enterprise AI use cases, identifying high-severity LLM vulnerabilities
- Designing scalable workflows for AI security assessments, reporting, and governance
- Evaluating state-of-the-art AI security tools and coding assistants from a security perspective
- Contributing to internal AI security research and emerging threat analysis for Generative AI systems
Recognition: Inventor Recognition (Q4 2025) for filed patents · Speaker at DEVUP 2026 (JPMC's invite-only technical conference) · SEP Engineer Committee Lead for 1,100+ early-career engineers at JPMC Bengaluru Tech Centre
prompt-injection-ctf — Interactive AI Security Playground. Craft attack prompts to break constrained AI systems. Covers prompt injection, jailbreaking, intent drift & token smuggling. Built to teach adversarial thinking hands-on.
llm-ops-workshop — End-to-end MLOps workflow demonstrating model lifecycle, monitoring, and deployment practices.
ML-101-Workshop — Source code from the ML-101 workshop hosted by IEEE Bangalore Section at IEEE CCONNECT. Built to make machine learning accessible to early-career engineers.
schmaltz-surveyor — Live sentiment analysis of public tweets. Two-phase project: classifier benchmarking across multiple ML models, then a web app for real-time Twitter sentiment analysis.
Also building something in AI security — stealth mode 🔒
LLMs & AI Platforms
Red Team & Security Tools
Frameworks & Standards
Tech Stack
