Model Engineer, Post-Training & RL (Equity + Impact)

Maven Ventures; Palo Alto (CA); In loco; USD 180.000 - 600.000; Tempo pieno

Vantaggi offerti da questo lavoro

Equity

Medical, Vision, and Dental

401(k) retirement plan

Disability insurance

Descrizione del lavoro

SpaceXSI in Palo Alto seeks a Model Engineer, Post-Training & RL, to tackle challenging post-training and reinforcement learning tasks on the Grok stack. You will work on reward modeling, preference optimization, and RL to improve model reasoning and real-world capabilities.

You should be proficient in Python and ML frameworks (PyTorch, JAX) and comfortable with C++/Rust. The role comes with competitive compensation, equity, and comprehensive benefits as part of SpaceXSI's total rewards package.

Competenze

• Production-grade Python/PyTorch/JAX/Rust/C++ coding experience.
• Experience with post-training and RL techniques is a plus.
• Interest in truth-seeking AI and model alignment.

Mansioni

• Advance post-training and RL challenges including RLHF and preference optimization.
• Improve reasoning, truthfulness, and real-world capabilities of models.
• Collaborate across teams to ship reliable AI systems.

Conoscenze

Python

PyTorch

JAX

Rust

C++

Descrizione del lavoro

SpaceXSI in Palo Alto seeks a Model Engineer, Post-Training & RL, to tackle challenging post-training and reinforcement learning tasks on the Grok stack. You will work on reward modeling, preference optimization, and RL to improve model reasoning and real-world capabilities.

You should be proficient in Python and ML frameworks (PyTorch, JAX) and comfortable with C++/Rust. The role comes with competitive compensation, equity, and comprehensive benefits as part of SpaceXSI's total rewards package.