Member of Technical Staff (Inference)

Join Hyperbolic as an Inference Engineer, where you'll build and optimize inference capabilities on our unified control plane, Forge. You'll be responsible for deploying and serving models across global clusters, evaluating inference frameworks, and ensuring production readiness. This role requires a strong background in inference performance, Kubernetes, and hands-on model optimization.

Experience deploying inference across heterogeneous accelerators

About Hyperbolic

Responsable de la construction des capacités d'inférence sur Forge, le plan de contrôle unifié, afin que les clients puissent consommer des jetons de modèle sans gérer les GPU.

Responsibilities

• Propriétaire de la manière dont les modèles sont déployés et servis à travers des clusters distribués dans le monde entier, sur du matériel hétérogène.
• Évaluation des frameworks d'inférence, mise en place de la surveillance, des passerelles et des points de terminaison qui rendent un déploiement prêt pour la production.

Requirements

• Working knowledge of NVIDIA Dynamo and how it fits into a distributed serving architecture
• Generalist instincts: you're willing to pick up adjacent work when it's what the product needs
• Proven ability to build a product end to end — you've taken something from nothing to serving real traffic
• Strong self-initiative and comfort operating as the primary owner of an area with minimal direction
• Experience setting up monitoring, gateways, and endpoints for production inference services
• Familiarity with modern inference frameworks and serving engines, and the judgment to evaluate and select among them for a given workload
• Deep Kubernetes experience, including hands-on ability to operate clusters in production, not just deploy to them
• Solid grasp of the concepts that govern inference performance: TTFT, disaggregated inference, speculative decoding, and KV cache and its inner workings
• Strong general inference background with a broad, high-level command of the stack rather than a narrow specialty — you can reason about the whole path from request to token
• Experience spanning both inference deployment and inference optimization
• Hands-on model optimization work — quantization, batching strategies, kernel-level tuning, or similar
• Understanding of RDMA and high-performance networking as they apply to distributed serving
• Background supporting customers directly on inference debugging and performance issues
• Experience at a GPU cloud, inference provider, or AI infrastructure company