Principal AI/ML Engineer - AI Safety & Evaluation

Location: San Jose, California; hybrid

Targeted compensation: $225,000–$245,000 per year

Company

hirify.global provides the organizational setting for work focused on making AI systems safe, aligned, and robust against misuse.

What you will do

• Architect and lead model-level defenses against jailbreaks, prompt injection, custom policy violations, and harmful outputs.
• Define adversarial testing, evaluation, and stress-testing strategies to identify safety weaknesses before deployment.
• Set technical direction for safety-focused fine-tuning, prompt shielding, and post-processing mitigation techniques.
• Collaborate with red teamers and researchers to turn emerging threats into measurable evaluations and system safeguards.
• Scale human-in-the-loop pipelines for detecting toxic, biased, and non-compliant outputs.
• Apply current LLM safety research and adversarial trends to production defenses.

Requirements

• 7+ years of experience in applied machine learning, AI infrastructure, or safety-critical systems, including 3+ years in a senior or staff-level technical leadership role.
• Deep understanding of transformer-based architectures and experience building or evaluating LLM safety interventions.
• Expertise in analyzing and addressing adversarial behavior, edge-case failures, and misuse scenarios.
• Ability to guide long-term technical strategy, influence organizational direction, and mentor cross-functional teams.
• Strong written and verbal communication skills with experience influencing technical direction at the organizational or platform level.
• Bachelor’s, Master’s, or PhD in Computer Science, Machine Learning, or a related field.

Nice to have

• Experience with reinforcement learning from human feedback, adversarial training, or safety fine-tuning at scale.
• Hands-on experience designing prompt-level defenses, content filtering systems, or jailbreak prevention mechanisms.
• Contributions to AI safety research, industry standards, or open-source tools for robustness, alignment, or evaluation.
• Familiarity with model governance frameworks, safety policies, model cards, red teaming protocols, or risk classification.

Culture & Benefits

• Work at the intersection of applied research, systems engineering, model behavior, and product safety.
• Build safety systems intended for real-world production deployment.
• hirify.global is an equal opportunity employer and a VEVRAA federal subcontractor.

Hiring process

• Live interviews are designed to assess the candidate’s own skills and thinking.
• AI or recording tools are not permitted during live interviews unless explicitly invited or approved in advance as a reasonable accommodation.