Latest AI JobsBrowse Jobs
Handshake logo

AI Red Teamer (Seattle)

HandshakeSeattle, WAHAI Delivery Ops
Apply on company site
Salary
$32 – $95 per hour
Job type
Full-time
Experience
Mid
Posted
3 Oct 2026
Status
Open · checked 7d ago

About this role

Handshake seeks an AI Red Teamer for a full-time, hourly position based in Seattle, Washington. This role suits professionals with strong adversarial problem-solving skills who want to stress-test large language models for leading research labs.

Role Overview

The AI Red Teamer intentionally breaks large language models to expose vulnerabilities, unsafe content, biases, and prompt injection weaknesses. Operating through TCWGlobal as a W-2 employee, this position involves 40 hours per week in the Seattle metro area or requires relocation prior to the start date.

Key Responsibilities

Design creative prompts and multi-turn scenarios to test AI guardrails across various risk categories.
Bypass safety filters and restrictions using jailbreak, evasion, and prompt injection methods.
Investigate edge cases to generate disallowed, incorrect, or harmful outputs.
Score and evaluate model responses according to structured harm taxonomies and severity rubrics.
Record experiments clearly by noting attempts, strategies, and resulting discoveries.
Assess and improve adversarial prompts created by team members.
Help develop harm taxonomies, calibration exercises, and reliability tasks.
Share findings with engineers, researchers, and data scientists to improve system defenses.
Handle potentially disturbing content regularly.
Track current attack methods, jailbreaks, and changing model behaviors.

Requirements

Hands-on experience working with multiple large language models such as ChatGPT, Claude, Gemini, and open-source models.
Intuition for building adversarial prompts, with familiarity in evasion or jailbreak techniques considered a strong plus.
Adversarial and creative problem-solving capabilities.
Clear written communication skills.
Strong ethical judgment to separate adversarial testing from personal values.
Self-directed and collaborative work style suited for feedback-heavy settings.
Persistence, curiosity, and comfort with experimental failure.
Mandatory residency in the Seattle metro area or willingness to relocate before starting.

Nice to Have

Knowledge of Python or alternative scripting languages.
Background using LLM APIs or evaluation tools.
Experience in rubric-based scoring and structured data annotation.
Past experience in security research, content moderation, trust and safety, or quality assurance.
Subject matter expertise in high-risk domains like law, medicine, chemistry, biology, cybersecurity, or finance.

Skills and Tools

Large language models (ChatGPT, Claude, Gemini, open-source models)
LLM APIs and evaluation tooling
Python and other scripting languages
Prompt injection, jailbreak, and evasion techniques
Harm taxonomies and severity rubrics

Salary and Benefits

Hourly pay ranging from $32 to $95 per hour.
W-2 employment through TCWGlobal.
40 hours per week.

About the Company

Handshake AI collaborates with prominent AI research labs to enhance model safety and robustness. The company conducts red teaming operations to discover vulnerabilities prior to public release, supporting the responsible creation of frontier artificial intelligence systems.

How to Apply

Interested candidates should apply on the official company careers page.

Skills and tech stack

ChatGPTClaudeGeminiPythonLLMLLMs

Questions candidates ask

Where is the AI Red Teamer position located?

The role is located in Seattle, WA. Candidates must reside in the Seattle metro area or be willing to relocate prior to their start date.

What is the employment type and pay rate for this role?

It is a W-2, hourly position employed through TCWGlobal for 40 hours per week, with a salary range of $32 to $95 per hour.

What kind of content will I be exposed to in this job?

The role involves regular and deliberate exposure to harmful content, including violence, self-harm, hate speech, sexually explicit material, and child safety scenarios as part of structured adversarial testing.

What qualifications are needed for the role?

Requirements include hands-on experience with multiple LLMs, intuition for adversarial prompts, creative problem-solving skills, clear written communication, strong ethical judgment, and a self-directed, collaborative approach.

What experience with programming languages is helpful?

Familiarity with Python or other scripting languages is listed as a nice-to-have qualification.

Ready to apply?

The application is on the official Handshake careers page.

Apply on company site