Sonder Labs seeks a Research Engineer to enhance model capabilities through data pipelines, training environments, and reinforcement learning. The role involves building high-quality data systems, developing multi-step task environments, and improving post-training methods with rigorous experimentation. Candidates must possess strong Python engineering skills, experience with PyTorch/JAX, and deep technical knowledge of RL and policy optimization.
Key Highlights
Build high-quality data pipelines for generating, curating, and validating training data.
Develop reliable training environments for multi-step tasks with reproducible execution.
Improve post-training via supervised baselines, RL experiments, and trustworthy evaluations.
Key Responsibilities
Build high-quality data pipelines by developing systems for generating, curating, and validating training data.
Develop training environments for multi-step tasks with reproducible execution and useful feedback.
Improve post-training by establishing supervised baselines, running RL experiments, and investigating data and reward effects.
Make evaluations trustworthy by designing graders, investigating reward hacking, and distinguishing capability gains from benchmark artifacts.
Close the experimental loop by inspecting model failures, forming testable hypotheses, and using controlled ablations.
Technical Skills Required
Python
Reinforcement Learning
PyTorch
Discover our full range of relocation jobs with comprehensive support packages to help you relocate and settle in your new location.
Benefits & Perks
Comprehensive health, dental, and vision coverage
Flexible PTO
Relocation support
Meaningful equity
Job Description
Sonder is an applied AI lab building models that learn how people work.We bring research, engineering, and design together to build useful personal AI, with privacy and efficiency at the core. We are building our early team in New York.The RoleWe are looking for a research engineer to improve model capabilities through better data, training environments, and reinforcement learning.You will work across data generation, environment design, and post-training, with ownership from an initial hypothesis to a measured improvement in model behavior. The work calls for strong engineering, careful experimentation, and judgment about what is worth pursuing.
Want the full job description?
Read the complete details on LinkedIn, the original posting.
Continue on LinkedIn
This is a short excerpt. All rights to the full description belong to its original publisher.
