Junior Research Scientist (AI Evaluation)

About Advai

Advai partners with businesses, governmental bodies, and defense sectors to rigorously assess AI systems prior to their deployment. As AI integration accelerates across diverse industries, our goal is to empower organizations with evidence-based testing that ensures AI is implemented safely and reliably.

Role Summary

We are seeking a proactive Junior Research Scientist for AI evaluation who will directly collaborate with clients to analyze their AI technologies. This individual will conceptualize, construct, and execute tests aimed at validating AI safety before deployment, spanning commercial, public, and defense domains. The position combines scientific inquiry, engineering, and consultative engagement with a tangible impact on AI adoption.

Key Responsibilities

• Engage with customers to understand and onboard their AI use cases into Advai's testing platform.
• Clarify client risks and testing requirements, transforming abstract concerns into precise, testable criteria.
• Develop detailed test plans comprising datasets, scoring methods, and procedures to evaluate AI systems effectively.
• Implement and execute these tests on the Advai platform and generate comprehensive results.
• Collaborate with engineering teams to integrate testing workflows into the platform and contribute to user interface design for clear results communication.
• Provide expert advisory services, including governance and risk management guidance related to AI assurance.
• Analyze and interpret test outcomes and recommend subsequent steps for clients’ AI development journeys.
• Participate in internal research initiatives to advance Advai's automated AI testing and monitoring technologies.

Required Qualifications and Skills

• Bachelor’s or Master’s degree in Machine Learning, Computer Science, or a closely related discipline.
• Practical experience programming in Python and familiarity with AI/ML frameworks like PyTorch and Huggingface.
• Strong enthusiasm for AI testing, evaluation, and real-world AI application assessments.
• Effective communication skills for interacting directly with clients and articulating technical concepts clearly.
• Capability to operate across scientific test design, software development, and customer-facing roles.
• Keen curiosity and aptitude for rapid learning in emerging research areas.
• Excellent teamwork abilities, collaborating with both research and engineering colleagues.
• Proficient written and verbal communication skills for technical documentation and client presentations.

Preferred Attributes

• Experience with adversarial AI, red teaming, robustness testing, or AI security assessment.
• Familiarity with Linux operating environments and remote computing setups.
• Knowledge of software engineering processes including version control and testing frameworks.
• Background in advising on governance, risk, or assurance frameworks.

Benefits

• Flexible work schedule supporting hybrid engagement with at least two days a week onsite in London.
• Private healthcare coverage and a cycle-to-work benefit.
• Opportunities for continuous professional growth in the evolving AI testing field.
• Dynamic work environment dedicated to helping organizations deploy trustworthy AI solutions.

Diversity and Inclusion

Advai is committed to nurturing an inclusive culture and strongly invites applications from individuals of all backgrounds.