Senior Post-Training & RL AI Engineer

SpaceXAI; Palo Alto (CA); On-site; USD 180,000 - 600,000; Full time

Benefits offered by this job

Equity

Medical Insurance

Vision and Dental

401(k)

Disability Insurance

Life Insurance

Employee Discounts

Job summary

SpaceXSI is seeking a Sr. Model Engineer, Post-Training & RL to tackle post-training and reinforcement learning challenges, including reward modeling and RLHF/DPO to improve reasoning and truthfulness. You will develop capabilities across the Grok stack and applications.

You will work on models with 1B+ parameters, applying production-quality Python, PyTorch, JAX and C++ code to push the boundaries of AI performance and reliability.

Qualifications

• Production-quality Python, PyTorch, JAX, Rust or C++ code experience.
• Experience training large models (1B+ parameters) incl. post-training and RL.

Responsibilities

• Tackle post-training and RL challenges for model alignment and usability.
• Work on reward modeling and preference optimization (RLHF/DPO) to improve reasoning.
• Enhance real-world capabilities of Grok models and related AI stack.

Skills

Python programming

Machine learning

RL fundamentals

Post-training ML

Tools

Python

PyTorch

JAX

Rust

C++

Job description

SpaceXSI is seeking a Sr. Model Engineer, Post-Training & RL to tackle post-training and reinforcement learning challenges, including reward modeling and RLHF/DPO to improve reasoning and truthfulness. You will develop capabilities across the Grok stack and applications.

You will work on models with 1B+ parameters, applying production-quality Python, PyTorch, JAX and C++ code to push the boundaries of AI performance and reliability.