Senior Model Training & Scaling Engineer (Hybrid)

Intelix.AI; Seattle (WA); On-site; USD 147,000 - 220,000; Full time

Job summary

Intelix.AI is seeking a Senior Research Engineer on the model scaling team, reporting to the Director of AI Research. The role covers model training and the supporting infrastructure, with training runs on 600 B300 and 1,000 H100 GPUs in a hybrid Seattle environment.

Responsibilities include building the training stack for distributed training, running scaling experiments, and specializing in CUDA kernels, Mixture-of-Experts, or RL post-training, plus releasing models and code publicly.

Qualifications

• 6+ years of software engineering experience.
• 4+ years of machine-learning infrastructure experience.
• Experience training models from scratch (pretraining) is required; retrieval-augmented generation or fine-tuning only does not meet this.

Responsibilities

• Build and maintain the training stack for distributed model training across GPU clusters.
• Run scaling experiments covering model architecture, data, and optimisation.
• Specialise in one area: CUDA kernels, Mixture-of-Experts, or reinforcement-learning post-training (GRPO, PPO, RLVR).
• Release models, code, and associated research publicly.

Skills

Python

PyTorch

CUDA

Mixture-of-Experts

Reinforcement Learning

Tools

JAX

Job description

Intelix.AI is seeking a Senior Research Engineer on the model scaling team, reporting to the Director of AI Research. The role covers model training and the supporting infrastructure, with training runs on 600 B300 and 1,000 H100 GPUs in a hybrid Seattle environment.

Responsibilities include building the training stack for distributed training, running scaling experiments, and specializing in CUDA kernels, Mixture-of-Experts, or RL post-training, plus releasing models and code publicly.