Advance Artificial Intelligence as a Safety and Alignment Research Engineer at Character.AI

Join the forefront of AI advancement at Character.AI and contribute to the development of safe, aligned AI technologies. Our growing Post-Training team is looking for a talented Research Engineer in Safety and Alignment to enhance our cutting-edge AI models. This role offers the unique opportunity to shape AI applications that are integrated daily into people's lives, ensuring their safety and alignment with human values.

About the Role

As a part of our dynamic Post-Training team, you'll play a crucial role in evolving our state-of-the-art pretrained language models into fully optimized digital products. This role demands a proactive approach to cross-functional collaboration, improving model performance, and directly influencing the users' interaction experience, impacting millions daily.

Responsibilities include:

• Developing algorithms to align model preferences.
• Creating classifiers to detect and mitigate model failures and misuse.
• Collaborating with annotators and red-teamers to refine alignment datasets.
• Devising innovative strategies to guide AI behavior.

Job Requirements

We are looking for a candidate who can write both production-facing and training code with clarity and efficiency. The ideal candidate will have:

• Proven experience in handling GPUs for training, serving, and debugging of models.
• Expertise in managing data pipelines and infrastructure.
• A strong grasp of modern machine learning techniques, including reinforcement learning and transformers.
• A history of exceptional research or innovative applied machine learning projects.

Nice to Have:

• Background in developing safety systems for user-generated content or consumer content platforms.
• Experience with LLM alignment.
• Contributions to academic journals or major conferences in machine learning or recommendation systems.

About Character.AI

Launched in 2021 by AI visionaries Noam Shazeer and Daniel De Freitas, Character.AI has quickly escalated to a prime position in the AI industry, achieving robust growth, unicorn status, and recognition such as Google Play's AI App of the Year.

We encourage diverse candidates to apply, including those who might not meet every requirement. At Character.AI, we believe in the unique value of every individual and encourage you to step forward. Join us to be part of our mission to empower the future with AGI!

Job Title: Research Engineer (Safety and Alignment), Post-Training
Company Name: Character.AI

Character.AI is committed to diversity and equal opportunity. We do not discriminate on the basis of race, religion, national origin, gender, sexual orientation, age, veteran status, or disability. Your unique perspectives are what fuel our innovation.

Your journey in shaping the AI-driven future starts here. Apply today!