Research Engineer / Scientist, Alignment Science
AnthropicSalary not specified
- Full Time
- Junior (1 to 2 years)
Not SpecifiedCompensation
Mid-level (3 to 4 years), Senior (5 to 8 years)Experience Level
Full TimeJob Type
UnknownVisa
AI & Machine Learning, Consumer SoftwareIndustries
Requirements
Candidates must hold a Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, or a closely related field. A strong publication record in top-tier conferences such as NeurIPS, ICML, or ICLR, with a focus on reinforcement learning, alignment, or generative models is essential. Extensive experience in implementing and optimizing large-scale training pipelines using PyTorch is required, along with a deep understanding of reinforcement learning techniques, particularly RLHF. Candidates should have experience with distributed training systems and large-scale experiments, as well as a proven track record in designing and implementing robust evaluation frameworks. Excellent communication skills and strong software engineering skills are also necessary.
Responsibilities
The Research Scientist will lead research initiatives in alignment and post-training methods for video generation models, focusing on improved quality, reliability, and adherence to human intent. They will design and implement supervised fine-tuning and reinforcement learning from human feedback (RLHF) pipelines for video generation models, develop robust evaluation frameworks to measure model alignment, safety, and output quality, and create and optimize data collection pipelines for human feedback and preferences. The role involves designing and conducting experiments to validate alignment techniques and their scaling properties, collaborating with cross-functional teams to integrate alignment improvements into the production pipeline, and mentoring junior researchers while fostering a culture of responsible AI development.
Skills
Reinforcement Learning
Generative Models
Supervised Fine-Tuning
RLHF
Evaluation Frameworks
Data Collection
Experiment Design
Collaboration
Ph.D. in Computer Science
Artificial Intelligence
Machine Learning
Genmo
AI tools for multimedia content creation
About Genmo
Genmo.ai specializes in providing AI tools for generating and editing multimedia content, including images, videos, and presentations. Users can upload images and animate specific parts, like transforming a static sky into a timelapse, or create entire movies by refining ideas, generating scenes, and selecting transitions. The platform caters to both individual content creators and businesses, operating on a subscription model with various service tiers. Genmo.ai differentiates itself by continuously enhancing its technology and focusing on user intent, ensuring that clients have powerful tools to realize their creative projects.
San Francisco, CaliforniaHeadquarters
N/AYear Founded
$29.2MTotal Funding
EARLY_VCCompany Stage
Consumer Software, AI & Machine LearningIndustries
1-10Employees