Decagon
Senior Security Engineer, Detection & Response
Decagon, San Francisco, California, United States, 94102
Security Engineer
Join Decagon's Security team to protect our AI-powered customer experience agents that handle millions of real customer interactions daily. You'll develop detection systems that identify threats without disrupting the natural conversation flow that makes our AI agents effective. This role focuses on data pipelines, LLM-powered detection query writing, and automated "Watchtower" components of our security stack. We work with some of the leading vendors in the security data space, and you'll take ownership of the system to build an industry-leading DNR team. Decagon's team is some of the best in the industry, so you'll work alongside a skilled and enthusiastic team. In this role, you will: Collaborate with engineering to build low-latency systems that detect prompt injection, jailbreak attempts, and social engineering attacks against customer experience agents without introducing conversation delays Develop ML models that identify surprising or unexpected access patterns in the product Create an incident response system that analyzes prior interactions: piece together what occurred, what data was accessed, and where that information may have gone Build APIs and webhooks that allow enterprise customers to integrate our security insights into their existing SOC and incident response workflows Continuously research and model new threat patterns specific to customer service AI, including account takeover attempts and information extraction attacks Your background looks something like this: 4+ years building production security or data pipeline systems Advanced proficiency in Python with experience in data pipelines, automation tooling, and code review for production web applications Track record of building detection systems that balance security with user experience Experience with real-time data processing using Kafka, Pulsar, or similar systems for analyzing data streams Prior experience with tools such as Splunk, Panther, RunReveal, or others Experience with anomaly detection, sequence modeling, and statistical analysis of user behavior patterns Experience with SOC 2, ISO 27001, GDPR, and other enterprise security requirements Even better: Proven experience leveraging LLMs or AI tooling for efficiency improvements Strong understanding of customer service workflows and business impact of security measures Skilled and motivated to take advantage of advanced reasoning and software development tools such as Cursor, Claude Code, and Gemini for personal productivity improvements and design leverage Benefits: Medical, dental, and vision benefits Take what you need vacation policy Daily lunches, dinners and snacks in the office to keep you at your best Compensation: $200K
$300K + Offers Equity
Join Decagon's Security team to protect our AI-powered customer experience agents that handle millions of real customer interactions daily. You'll develop detection systems that identify threats without disrupting the natural conversation flow that makes our AI agents effective. This role focuses on data pipelines, LLM-powered detection query writing, and automated "Watchtower" components of our security stack. We work with some of the leading vendors in the security data space, and you'll take ownership of the system to build an industry-leading DNR team. Decagon's team is some of the best in the industry, so you'll work alongside a skilled and enthusiastic team. In this role, you will: Collaborate with engineering to build low-latency systems that detect prompt injection, jailbreak attempts, and social engineering attacks against customer experience agents without introducing conversation delays Develop ML models that identify surprising or unexpected access patterns in the product Create an incident response system that analyzes prior interactions: piece together what occurred, what data was accessed, and where that information may have gone Build APIs and webhooks that allow enterprise customers to integrate our security insights into their existing SOC and incident response workflows Continuously research and model new threat patterns specific to customer service AI, including account takeover attempts and information extraction attacks Your background looks something like this: 4+ years building production security or data pipeline systems Advanced proficiency in Python with experience in data pipelines, automation tooling, and code review for production web applications Track record of building detection systems that balance security with user experience Experience with real-time data processing using Kafka, Pulsar, or similar systems for analyzing data streams Prior experience with tools such as Splunk, Panther, RunReveal, or others Experience with anomaly detection, sequence modeling, and statistical analysis of user behavior patterns Experience with SOC 2, ISO 27001, GDPR, and other enterprise security requirements Even better: Proven experience leveraging LLMs or AI tooling for efficiency improvements Strong understanding of customer service workflows and business impact of security measures Skilled and motivated to take advantage of advanced reasoning and software development tools such as Cursor, Claude Code, and Gemini for personal productivity improvements and design leverage Benefits: Medical, dental, and vision benefits Take what you need vacation policy Daily lunches, dinners and snacks in the office to keep you at your best Compensation: $200K
$300K + Offers Equity