Staff Attack Engineer, AI/LLM
Horizon3 · Remote · US, Remote ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Staff level role
- Fully remote
- AI/LLM Security focus
- Python expert required
- $235k-$275k base
Join Horizon3 as a Staff Attack Engineer specializing in AI and LLM security. You will research vulnerabilities in AI and agentic systems, translating these findings into automated, repeatable attack modules for the NodeZero autonomous pentesting platform. This role requires driving technical strategy for LLM-powered offensive capabilities rather than performing manual consulting or pentesting.
What you'll do
- Break AI and agentic systems to create automated attack modules for NodeZero
- Design prompt injection and defense evasion attacks with reusable patterns
- Exploit tool-use by abusing LLM agent access to code, APIs, and databases
- Target AI infrastructure including model serving, training pipelines, and vector databases
- Research model and supply chain attacks such as poisoning and data extraction
- Perform threat modeling for agentic systems to map trust boundaries and attack surfaces
- Build and extend LLM-powered applications with a focus on cost, safety, and reliability
- Design microservices that orchestrate LLM tasks and integrate with offensive workflows
What you'll need
- Expert-level Python and software engineering skills
- Solid penetration testing fundamentals and understanding of attack chains
- Familiarity with AI/LLM security frameworks like OWASP Top 10 for LLMs and MITRE ATLAS
- Experience in a security product or offensive security team with shipped tooling
- Proven ability to break AI/LLM and agentic systems
- Clear understanding of trust boundaries around AI tools, data sources, and permissions
- Expert-level ownership capable of driving high-complexity programs and setting strategy
- Self-motivated with proactive problem-solving abilities
- Deep knowledge of the fast-moving AI security landscape and recent developments
Nice to have
- Experience with cloud AI services like Azure OpenAI or GCP Vertex AI
- Contributions to AI security research via blogs, talks, CVEs, or open-source tools
- Experience with AWS Bedrock and AWS Agent Core
- Familiarity with graph databases such as Neo4j
- Background in traditional exploit development or vulnerability research
- CTF experience, particularly in AI/ML-focused challenge categories
Stack
- Python
- NodeZero
- LLMs
- AWS Bedrock
- Azure OpenAI
- GCP Vertex AI
- Neo4j
How you'll work
- 100% remote company
- Work-life balance supported by remote policy
What you get
- Base salary range $235,000 - $275,000 annually
- Equity package in the form of stock options
- Health, vision, and dental insurance for employee and family
- Flexible vacation policy
- Generous parental leave
- Inclusive culture valuing diversity and equity