Applied Research Engineer Category Computer/software Location San Francisco, California Job Type Full Time
hireVouch
San Francisco, California, USA
On-site
Reinforcement learning from human feedback (rlhf)
Direct preference optimization (dpo)
Human data quality measurement
Advance the field of AI alignment by developing cutting-edge methods, such as RLHF and novel approaches, that ensure AI systems reflect human preferences more accurately
Job Summary
Advance the field of AI alignment by developing cutting-edge methods, such as RLHF and novel approaches, that ensure AI systems reflect human preferences more accurately.
Improve the quality of human-in-the-loop data by designing and deploying rigorous measurement and enhancement systems, leading to more reliable AI training.
Increase efficiency and effectiveness in AI-assisted data labeling by creating tools that leverage active learning and adaptive sampling, reducing manual effort while improving accuracy.
Matching Summary
Advance the field of AI alignment by developing cutting-edge methods, such as RLHF and novel approaches, that ensure AI systems reflect human preferences more accurately.
Skills & Requirements
Must-have
Reinforcement Learning from Human Feedback (RLHF)
Direct Preference Optimization (DPO)
human data quality measurement
AI-assisted data labeling tools
large language models
multimodal models
Nice-to-have
impact over process
innovation at speed
continuous learning
clear ownership
publish in top-tier journals
Key Requirements
Ph.D. or Master’s degree
3+ years of experience
Proficiency in Python
experience with PyTorch, JAX, or TensorFlow
track record of publishing in top-tier AI/ML conferences