Staff Engineer, Multi-Modal Audio & On-Device AI

Liquid AI; San Francisco (CA); Hybrid; USD 170,000 - 260,000; Full time

Benefits offered by this job

Equity

Health insurance

401(k) matching

Unlimited PTO

Job summary

Liquid AI, based in San Francisco, seeks engineers to advance frontier Audio AI. You will ship production-grade code handling data pipelines, evaluation, and deployment for on-device, low-latency systems. This role offers ownership of critical workstreams in a tight, high-skill team.

You will collaborate across data, research, and infrastructure to push end-to-end ML solutions, including optimization for device constraints and real-world deployments.

Qualifications

• Experience shipping production ML systems including data pipelines, evals, and serving infra.
• Strong fundamentals in clean, maintainable code.
• Proficiency with PyTorch and distributed training frameworks.

Responsibilities

• Build and scale data pipelines for audio model training, including preprocessing, augmentation, and quality filtering at scale
• Design, implement, and maintain evaluation systems that measure multimodal performance across internal and public benchmarks
• Fine-tune and adapt audio models for customer-specific use cases, owning delivery from requirements through deployment
• Contribute production code to the core audio repository, collaborating with infrastructure and research teams
• Support experimentation under real hardware constraints, shifting between customer work and core development as priorities evolve

Skills

Production-grade code

ML systems

PyTorch

Distributed training

Tools

DeepSpeed

FSDP

Git

Job description

Liquid AI, based in San Francisco, seeks engineers to advance frontier Audio AI. You will ship production-grade code handling data pipelines, evaluation, and deployment for on-device, low-latency systems. This role offers ownership of critical workstreams in a tight, high-skill team.

You will collaborate across data, research, and infrastructure to push end-to-end ML solutions, including optimization for device constraints and real-world deployments.