Launch-Focused ML Engineer – Model Integrations

Nunchux AI; San Francisco (CA); Hybrid; USD 170.000 - 240.000; Vollzeit

Benefits dieser Stelle

Hugging Face / Diffusers exposure

External model providers collaboration

Open-source ML contributions

Zusammenfassung

Nunchux AI is building infrastructure to speed and reduce the cost of serving multimodal generative AI. You will lead the initial integration of new image and video models into Modelverse, ensuring a fast Day 1 release and smooth handoff for performance and cost optimizations.

You will work with cloud and product teams to deploy, integrate APIs, and document launch requirements, while maintaining reusable adapters for future model integrations.

Qualifikationen

• Image or video model experience: you have run and adapted generation models or integrated provider APIs.
• Python and PyTorch: strong in both and able to read model code and debug pipelines.
• Production systems experience: deployed a model or built an API integration and debug failed requests.
• Working style: quick to learn unfamiliar model code and ship a prototype quickly.

Aufgaben

• Prototype new models and get them running with available code and weights.
• Ship the first version by coordinating deployment with cloud teams and API integration with engineers.
• Speed up launch processes by automating steps and building reusable adapters and scripts.
• Verify end-to-end integration, fix blockers, and document known limitations for downstream teams.
• Maintain the integration platform between launches and extend support for new interfaces and modalities.

Kenntnisse

Image/video model experience

Python and PyTorch

Production systems experience

Fast prototyping

Jobbeschreibung

Nunchux AI is building infrastructure to speed and reduce the cost of serving multimodal generative AI. You will lead the initial integration of new image and video models into Modelverse, ensuring a fast Day 1 release and smooth handoff for performance and cost optimizations.

You will work with cloud and product teams to deploy, integrate APIs, and document launch requirements, while maintaining reusable adapters for future model integrations.