Member of Technical Staff - RL Inference

X AI
Palo Alto
Workplace: OnsiteFull timeFunction: Software EngineeringSkills: ["Communication","Initiative","Prioritization","Problem-solving"]

Build and optimize an RL inference stack for low-precision training and inference at scale, from small experiments to production training runs. Profile and resolve performance bottlenecks in large distributed RL systems, and collaborate with the modeling team to implement novel RL techniques and algorithms efficiently. Work across the stack while focusing on efficiency, numerics, and quantization for LLM inference and training.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
X AI
X AI
1 month ago

Member of Technical Staff - RL Inference

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 9 hours agoStatus: Live

Job Summary

Build and optimize an RL inference stack for low-precision training and inference at scale, from small experiments to production training runs. Profile and resolve performance bottlenecks in large distributed RL systems, and collaborate with the modeling team to implement novel RL techniques and algorithms efficiently. Work across the stack while focusing on efficiency, numerics, and quantization for LLM inference and training.
Location: Palo Alto
Workplace: Onsite
Employment Type: Full time
Job Function: Software Engineering

Key Responsibilities

  • •Design and optimize the inference stack for RL workloads, from small ablations to production training runs.
  • •Analyze, profile, and address performance bottlenecks in large-scale RL systems.
  • •Collaborate with the modeling team to efficiently implement novel RL techniques and algorithms.

Key Requirements

  • •Experience building, debugging, and optimizing efficiency of large-scale distributed systems.
  • •Experience in LLM inference.
  • •Proficiency in Python, C++, and/or Rust, with frameworks such as PyTorch, Jax, and CUDA.
  • •Willingness to dive deep and solve hard problems across all levels of the stack.
Experience:LLM inference
Skills:CommunicationInitiativePrioritizationProblem-solving
Languages:English
Tech Stack:PythonC++RustPyTorchJaxCUDASGLangVLLM

Company Brief

X AI
Develops advanced artificial intelligence models and research aimed at building safe, general AI and understanding the fundamental nature of the universe. Focuses on large-scale AI systems, research publications, and building foundational AI capabilities.
Industry: AI & Machine Learning
Company Size: Medium (51 to 250 employees)
Growth: Early Stage Startup
Headquarters: San Francisco, United States
Founded: 2023
Website