Security Engineer, Anti-Abuse

About the Company

Arena is the platform for evaluating how AI models perform in the real world. Founded by researchers from UC Berkeley's SkyLab, we're on a mission to measure and advance the frontier of AI for real-world use, and to build the foundation for everyone to understand, shape, and benefit from it.

Responsibilities

• Own the abuse vision for Arena, including detection, enforcement, and false-positive management
• Design and operate detection for bots, sybils, coordinated inauthentic voting, and rating-system manipulation
• Build enforcement primitives such as rate limits, challenges, shadowbans, and account actions
• Detect and mitigate inference abuse and cost exploitation at the platform layer
• Build jailbreak and multi-provider misuse detection across served models
• Scope and implement abuse monitoring for all new product launches
• Prototype and mature production systems for high-severity harms and legal reporting pipelines
• Build internal investigator tooling for policy and T&S analysts
• Partner with Security on account takeover, credential stuffing, and API-key abuse
• Partner with policy, legal, and leadership on acceptable-use policy and enforcement

Requirements

• 6+ years of production software engineering experience building systems under adversarial conditions
• Shipped experience in trust & safety, anti-abuse, anti-fraud, anti-spam, integrity, or risk engineering
• Strong SQL and data-analysis skills
• Adversarial and investigative mindset
• High judgment on false-positive costs and user harm
• Proficiency in Node.js, TypeScript, Python, or Go
• Excellent communication skills

Preferred Qualifications

• Experience with LLM-specific adversarial inputs like jailbreaks and prompt injection
• Experience with agent safety, browser-automation abuse, or LLM-API abuse
• Background in securing voting, rating, reputation, or marketplace platforms
• ML or ML-systems experience including feature engineering and evaluation
• Experience building investigator or analyst tooling for non-engineers
• Contributions to open-source trust & safety or adversarial-ML work
• Background in gaming integrity, ad-fraud, or financial-crime engineering