Hugging Face Benefits Overview

Own Hugging Face’s open-source speech-to-speech library and build hf-voice into a production voice-agent platform. Responsibilities include architecture, realtime inference, ASR/TTS/model integration, streaming APIs, WebSockets/WebRTC, GPU serving, concurrency, autoscaling, observability, reliability, load testing, documentation, community contributions, and deployment support. The role requires substantial Python open-source experience, realtime systems expertise, production LLM or multimodal model experience, and interest in conversational AI.

Work on the Xet Storage team building and operating the Xet storage backend and xet-core (Rust) and hf-xet (Python). Design, implement, and operate high-performance, scalable storage systems, contribute to open-source, and support production monitoring, alerting, and distributed debugging in a remote, fast-moving environment.

Work on Xet Storage: contribute to xet-core (Rust) and the Xet backend to design, build, and operate high-performance, large-scale storage systems (200+ PB). Implement features, maintain reliability, and operate production services in a remote, low-process environment while contributing to open-source projects.