Odyssey
Member of Technical Staff, Foundation Models
Palo Alto +2 · Staff · Full-time · Posted Jul 24, 2026
- Location
- Palo Alto
Zurich
London - Experience
- 2+ years of software engineering experience and 4+ years of ML engineering experience, or a PhD in a related field.
- Education
- PhD in a related field is an alternative to 4+ years of ML engineering experience.
About the job
Build and train large, real-time foundation world models, owning the work end to end from data and architecture through inference and evaluation.
What you'll do
- Run scaling studies and use scaling laws to guide large training runs.
- Own distributed model training at scale, including data mixtures, loss and stability work, and performance improvements.
- Implement ML algorithms and work across pipelines, tokenizers, real-time inference, evaluation, and core ML frameworks.
What they ask for
- At least 2 years of software engineering experience, with significant work in ML performance.
- At least 4 years of ML engineering experience, or a PhD in a related field.
- Hands-on experience training large models and pretraining at multi-node scale, including debugging.
- Proficiency with PyTorch or TF/JAX.
Tools
PyTorch, TF, JAX
AI focus
Train and scale causal, multimodal world models that predict and interact with the world over long horizons.
Model-written summary · the original posting is authoritative