Sanket Mendapara

AI Researcher

Seattle, Washington, United States6 yrs 3 mos experience

Highly StableAI Enabled

Key Highlights

Expert in AI security and threat research.
Developed open-source tools for AI threat detection.
Led AI red teaming initiatives at Cisco.

Stackforce AI infers this person is a Cybersecurity expert specializing in AI security and threat research.

Contact

Skills

Core Skills

Ai SecurityThreat Research

Other Skills

Agentic AI SecurityDetection EngineeringAnti-spam TechnologiesPythonBot DetectionLinuxComputer NetworkingRouting ProtocolsAI Threat IntelligenceNatural Language UnderstandingNatural Language Processing (NLP)Responsible AIAI Safety EvaluationLLM BenchmarkingAI Firewall

About

🔬 Research Areas • Threat research and intelligence analysis, AI threats and risks frameworks, LLM and agentic system security (including MCP/A2A, skills security, etc.), Runtime security and guardrails, Multimodal AI security, AI supply chain security, including model security and provenance, Adversarial evaluation and automated red teaming, Detection Engineering & Endpoint AI security, Mechanistic interpretability 🧩 Previous Work Malware analysis, spam analysis, reverse engineering, and large-scale abuse detection. Openreview: https://openreview.net/profile?id=~Sanket_Mendapara1 Google Scholar: https://scholar.google.com/citations?hl=en&user=R60tspkAAAAJ

Experience

6 yrs 3 mos

Total Experience

3 yrs 5 mos

Average Tenure

2 yrs 10 mos

Current Experience

Cisco

AI Security Research Engineer

Jul 2023 – Present · 2 yrs 10 mos · Seattle, Washington, United States · Remote

Working as part of Cisco AI’s Threat Research & Security team, contributing to AI security threat research, adversarial analysis, and protection strategies across Cisco’s AI Security and AI Defense product ecosystem.
Co-building and maintaining the Cisco AI Security Framework (https://learn-cloudsecurity.cisco.com/ai-security-framework), a comprehensive AI threat and risk taxonomy covering LLM, agentic AI, and Model Context Protocol (MCP) ecosystems, evolved through real-world attack research.
Developing and maintaining open-source and enterprise AI security tools, including MCP-Scanner (https://github.com/cisco-ai-defense/mcp-scanner) and A2A-Scanner (https://github.com/cisco-ai-defense/a2a-scanner), enabling automated discovery and analysis of AI threats and vulnerabilities across LLM-, agentic-, and protocol-based (MCP/A2A) systems.
Designing and executing AI red teaming workflows, leveraging adversarial techniques such as jailbreaks, prompt injection, obfuscation, PII leakage, and harmful content generation to stress-test AI systems.
Building and maintaining AI runtime security detections, using signature-based and behavior-driven detection engineering to protect against AI-native and LLM-specific threats.
Driving automated AI red teaming, attack-path analysis, and combination-risk initiatives, enabling continuous security assessment and correlated risk visibility across AI infrastructure.
Collaborating closely with ML, Product, Engineering, and Threat Research teams to integrate AI guardrails, safety mechanisms, evaluation pipelines, and detection logic into Cisco’s AI products.
Previously served as Threat Detection Lead for Armorblox Email Protection, contributing to detection strategy and adversarial coverage during and after the Cisco acquisition.

AI SecurityAgentic AI SecurityThreat Research

Armorblox (now part of cisco)

2 roles

Threat Researcher

Jun 2023 – Jul 2023 · 1 mo

Working towards the advancement of Email Security and generic Threat Research!

Anti-spam TechnologiesPythonThreat Research

Threat Research Intern

Jan 2023 – May 2023 · 4 mos

Anti-spam TechnologiesPython

Cequence security

Threat Research and Engineering Intern

May 2022 – Aug 2022 · 3 mos

Worked with CQ-Prime team

Bot DetectionPython

Sophos

2 roles

Threat Researcher 2

Jul 2018 – Jul 2021 · 3 yrs

Worked on analysis and reverse engineering Portable Executable (PE) files and Non-PE files - PDF, HTML, JAR and scripting files for the latest Malwares and PUAs
Tracked and blocked the different attack vectors and Indicators of Compromise (IoCs) used in Zero-Day exploits and multi-staged malwares spread through spam emails
Working with Web Categorization Team on filtering & classification of different types of URLs/IPs into malware, phishing, CnC
Developed generic & heuristic detections to block different widespread spam campaigns
Trained new-recruits in advance spam analysis and the latest anti-spam detection techniques for the past 2 years
Handle FPs from customers triggered in anti-spam, URL filtering or anti-malware filtering layers and FP auditing of existing spam detections

Anti-spam TechnologiesLinuxThreat Research

Threat Research Intern

Jan 2018 – Jun 2018 · 5 mos

Spam Analysis

Anti-spam TechnologiesLinux

Microsystem solutions

Intern

May 2016 – Jul 2016 · 2 mos · Jaipur Area, India

Implemented various projects like - Designing and configuring various LAN and WAN scenarios on Cisco Routers - 2500, 2800 series and Switches - Catalyst 2950, 2960, 3550, 3560.
Worked on various protocols in LAN and WAN Implementation like Ethernet, Fast Ethernet, Gigabit Ethernet, Telnet, VTP, STP, PVST, RPVST HSRP, VRRP, OSPF, OSPF V-3, RIP, RIP-V-2, EIGRP,DHCP, DNS, Frame relay, PPP, HDLC, GRE, IPSec, etc. Planning and implementing different types of ACLs for small and mid-sized business network scenarios. Implemented NAT and Route Maps concepts on these business networks.

Computer NetworkingRouting Protocols