AI Red Teamer, Cybersecurity (Remote)
Handshake · Remote · Remote (USA) ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Fully remote (US)
- Part-time available
- Offensive security focus
- AI safety evaluation
- Senior level role
Join Handshake AI to test whether large language models can be manipulated into generating functional malware or attack guidance. You will simulate threat actors to identify gaps in safety guardrails and help secure frontier AI systems.
This position supports researchers at leading AI labs by solving complex data challenges. You will work with a distributed team of engineers and scientists to refine evaluation frameworks and improve model robustness.
What you'll do
- Design adversarial prompts to test model assistance across the cyber kill chain
- Evaluate generated code for functional correctness and real-world exploitability
- Test offensive categories like malware generation and privilege escalation
- Simulate attacker personas ranging from opportunistic to APT-level actors
- Probe dual-use boundaries between legitimate research and offensive applications
- Score responses using structured harm taxonomies and severity rubrics
- Document findings with clear technical reasoning on attacker assistance levels
- Collaborate with researchers to translate findings into model improvements
What you'll need
- Experience in offensive security, red teaming, or vulnerability research
- Ability to read and write code in Python, PowerShell, Bash, C/C++, or JavaScript
- Understanding of MITRE ATT&CK, OWASP, and common attack frameworks
- Hands-on experience using multiple LLMs like ChatGPT, Claude, or Gemini
- Skill in assessing functional correctness of model-generated technical output
- Strong ethical judgment and ability to separate adversarial thinking from values
- Clear written communication skills for explaining technical risk
- Ability to work independently in a distributed, feedback-heavy environment
Nice to have
- Certifications such as OSCP, OSCE, GPEN, or CEH
- Active or previous security clearance
- Experience with exploit development, reverse engineering, or binary analysis
- Background in cloud, container, or infrastructure-as-code security
- Familiarity with AI attack surfaces like prompt injection or model extraction
- Track record in bug bounties or published CVEs
- Previous work in trust and safety or AI evaluation
Stack
- Python
- PowerShell
- Bash
- C/C++
- JavaScript
- ChatGPT
- Claude
- Gemini
How you'll work
- Remote within the United States
- Optional presence in Seattle, WA
- Flexible work arrangement
- Part-time availability accepted
- Distributed team collaboration
What you get
- Partner with world-class AI labs and Fortune 500 companies
- Work with professionals from Palantir, Meta, and Scale AI
- Impact career evolution in the AI economy at global scale
- Contribute to a business with approximately $1 billion run rate