Member of technical staff (Safety & Data)

As a Member of Technical Staff in Safety & Data at H, your primary responsibilities include designing and executing model-level and system-level safety evaluations, contributing to the codebase for training and fine-tuning AI models, collaborating with cross-functional teams to implement best practices in responsible AI, and conducting research to identify and mitigate new safety risks specific to agentic AI.

For the Member of Technical Staff (Safety & Data) position at H, candidates are expected to have a PhD in machine learning, deep learning, or responsible AI. Essential skills include a deep understanding of safety mitigation strategies for L(V)LMs, proficiency in Python, and experience with deep learning frameworks like PyTorch or TensorFlow.

The work culture at H encourages openness, collaboration, and continuous learning. As a Member of Technical Staff in Safety & Data, you'll be part of a dynamic and multicultural team that values impactful communication and teamwork, making it an exciting environment for professional growth.

H prioritizes safety by implementing a robust culture of safety and risk mitigation at every development stage. The Safety & Data team is dedicated to proactively identifying risks, establishing guardrails, and conducting thorough qualitative and quantitative assessments to ensure the safety of its agentic AI offerings.

A Member of Technical Staff (Safety & Data) at H has numerous opportunities for career advancement, including continuous skills development, participation in innovative projects, and collaboration with leading experts in AI. The company's commitment to professional growth means you'll have the chance to advance while shaping the future of AI.

When answering this question, highlight specific projects where you've implemented safety evaluations, detailing what methods you used and the outcomes. Emphasize your understanding of both qualitative and quantitative assessments.

In your response, discuss past experiences working cross-functionally. Detail your communication methods and how you ensured productive collaboration, showcasing your team-oriented mindset while addressing safety challenges.

Share your knowledge on adversarial ML and approaches you've implemented. This could include specific tools or frameworks you used and examples of how you automated detection processes in your previous roles.

Discuss frameworks or checklists you use to ensure responsible AI practices. Highlight any relevant experiences where you had to address ethical considerations in your projects, demonstrating your commitment to ethical AI development.

Provide specific examples of models you've successfully fine-tuned, discussing the techniques employed and the improvements observed in performance metrics. Detail your proficiency with relevant deep learning frameworks.

Mention resources such as industry conferences, journals, or online courses that you utilize to keep your knowledge current. Emphasize your proactive approach to learning and how you apply new insights to your work.

Reflect on a real project, outlining the specific challenges encountered and the strategic steps taken to overcome them. Highlight what you learned and how the experience influenced your future work.

In your answer, list the programming languages you’ve used extensively, giving specific examples of how you've applied them in real-world projects. Mention your comfort level with various deep learning frameworks.

Discuss key performance indicators (KPIs) and metrics you track to assess the effectiveness of safety protocols. Provide examples where you have successfully measured and adjusted protocols based on those findings.

To answer this, speak to your passion for AI combined with ethical considerations. Mention H's mission and values, and express your enthusiasm for contributing to the development of safe, responsible AI technologies.