Member of Technical Staff, Pretraining Architecture and Scaling
Hark · San Jose · HQ
Frontier labonsite$180k–$450k
At a glance
Highlights
- Build natively multimodal foundation models for real-time Omni experiences.
- Research pretraining architectures and scaling to improve largest models.
- Develop unified intelligence pairing multimodal models with next-gen hardware.
Ideal candidate
Experimental researcher who loves scaling large models and thrives debugging complex training dynamics. You care about efficient, capable multimodal intelligence and turning rigorous experiments into production foundations.
Requirements
- Hands-on experience training large models balancing quality, compute cost, and stability
- Strong experimental judgment designing controlled experiments
- Experience applying scaling laws and evaluating extrapolations from smaller runs
- Experience implementing and debugging distributed training across GPU clusters
- Strong engineering skills profiling workloads and debugging model behavior
- Record of improving model quality or training efficiency
Lab profile
About Hark
Building proactive, personalized multimodal AI with next-generation hardware for human-machine interaction.
Known for
- Proactive personalized intelligence
- Multimodal speech, text and vision
- Persistent memory
- Next-generation AI hardware
Posted · Last seen