Innodata Inc is hiring a Generative AI Associate - Red Teaming Specialist. You will be at the forefront of ensuring AI safety by creatively challenging large language models (LLMs) to uncover weaknesses and improve system resilience. This role places you directly on a global team dedicated to making AI more robust and reliable.
What You'll Do
- Complete extensive training on AI/ML, LLMs, Red Teaming, jailbreaking, and specific project guidelines.
- Craft clever and sneaky prompts to attempt to bypass filters and guardrails on LLMs, targeting specific client-defined vulnerabilities.
- Collaborate closely with language specialists, team leads, and QA leads to produce high-quality work.
- Assist data scientists to conduct automated model attacks.
- Adapt to the dynamic needs of different projects and clients, navigating shifting guidelines.
- Keep up with the evolving capabilities and vulnerabilities of LLMs and help your team's methods evolve.
- Hit productivity targets, including number of prompts written and average handling time per prompt.
Team & Environment
You will join a global team of subject matter experts, working directly with language specialists, team leads, QA leads, and data scientists.
Work Mode
This is a local-country position open to candidates in Canada, excluding Quebec.
Innodata Inc is an equal opportunity employer.
