Headline

AI Agent Security Engineer | Creator of Pinaka.sh | DEFCON | BSides Luxembourg AI Security Village Host | 13+ Conferences

About

I research the security of AI agents and the new vulnerability classes that emerge when language models are given tool access and control over real infrastructure. It's an attack surface most security teams have not started thinking about yet: agentic risk, shadow AI, and the gap between how these systems are built and how they fail. I build Pinaka, an MCP-native autonomous agentic pentesting platform. It autonomously identified and reported a Critical-severity finding on RedBull's public bug bounty program — real vulnerabilities, real production targets, found on its own. At Harness I work on securing AI workloads and agentic deployments at enterprise scale. Previously at Cequence Security and Aikido Security, building API and AI security tooling. I have spoken at 13+ international security conferences including BSides, OWASP, AI Dev World, and API World. In May 2026 I hosted the AI Security Village at BSides Luxembourg — a 2-day track on offensive AI, agentic risk, and AI-as-a-Service security. If you are building or securing AI-native systems, let's talk.

Location

Bengaluru

AI stack

anthropic apis

claude code

hugging face

mcp

openai

xai

Work experience

Senior Security Research Engineer - AI Security at Harness, Nov 2025 - Aug 2026 · 10 mos, Bengaluru, - Working on AI Security and Agentic Security, focused on GenAI/MCP security, AI discovery, compliance, observability, and safe AI adoption. Built end-to-end security detections for MCP Tools, Prompts, and Resources, covering risks like sensitive data exposure, unauthenticated access, insecure exposure, lack of encryption, server/stack trace disclosure and covering OWASP MCP and LLM Top 10. - Built large-scale AI safety datasets and training pipelines for LLM risk detection, including prompt injection and harmful-content classification. - Researching on AI and MCP Attack vectors and building exploit.

Artificial Intelligence Engineer at Aikido Security, Sep 2025 - Nov 2025 · 3 mos, Built and deployed security AI agents for automated exploit testing, improving prompt logic and agent performance. Implemented scalable attack-surface detection workflows using AWS Lambda and SQS. Created real vulnerable environments from GitHub Advisories to validate and refine exploit capability of AI agents.

Security Engineer at Cequence Security, Jan 2024 - Sep 2025 · 1 yr 9 mos, San Francisco, CA, - Spearheaded the creation of advanced test suites for API Security Testing—fully aligning with OWASP API Top 10 and LLM Top 10 standards—to enable robust, real-world validation of modern software architectures. - Developed a scalable Attack Surface Management platform leveraging no-code automation, streamlining reconnaissance, accelerating vulnerability triage, and significantly minimizing manual workload. - Performed live API monitoring and conducted real-time threat audits, dramatically enhancing incident detection and response capabilities. - Automated critical security workflows, boosting test velocity, reducing manual effort, and elevating team-wide operational efficiency.

Security Analyst Intern at Cequence Security, Jan 2023 - Dec 2023 · 1 yr, Sunnyvale, CA, - Demonstrated proficiency in monitoring API activity and conducting routine security audits to promptly identify and address any potential security incidents. - Streamlined crawling processes and automated vulnerability discovery to enhance efficiency. - Proactively identified and reported API security vulnerabilities.

Student Network Technician at Northeastern University, Sep 2022 - Jan 2023 · 5 mos, Boston, MA, - Program and configure network switches to ensure optimal connectivity and performance across the organization. - Assist full-time staff with RMA (Return Merchandise Authorization) operations and provide proactive support for additional network tasks. - Maintain and enhance wireless infrastructure by replacing access points and conducting comprehensive site surveys to ensure seamless coverage. - Manage inventory lifecycle by deprecating outdated assets and executing accurate asset disposition procedures for compliance and efficiency.

Founder at BUG XS, Mar 2019 - Jan 2025 · 5 yrs 11 mos, • Founded and scaled a cybersecurity community—empowering 700+ students to identify real-world web vulnerabilities through hands-on learning. • Delivered expert-led training sessions on bug bounty hunting, producing multiple early-career success stories among mentees. • Organized and facilitated seminars and workshops at leading colleges, driving awareness and practical skill-building in application security.

Education

Master of Science - MS, Cybersecurity, Northeastern University

Bachelor’s Degree, Computer Science, Gujarat Technological University (GTU)

Public links

github: https://github.com/0xParth

youtube.com: https://youtube.com/@bugXS