Realtime Multimodal AI Engineer

Sable; San Francisco (CA); On-site; USD 150,000 - 230,000; Full time

Job summary

Sable is assembling a world-class team to build Aidan, the first AI employee capable of leading customer calls across voice, vision and browser use. You will empower Aidan to orchestrate real-time modalities—voice, hands, eyes, and the Brain—while refining the verification loop to ensure robust performance in live scenarios.

You will work in a small, autonomous team, delivering end-to-end engineering across perception, action planning, and continuous learning.

Qualifications

• Experience in building real-time AI systems.
• Proficiency with large language models and agent orchestration.
• Hands-on work across browser/UI automation.
• Applied ML for multimodal evaluation and improvement.

Responsibilities

• Improve runtime for voice latency, perception, and actions.
• Build Skills system from demos to durable knowledge.
• Enhance Aidan's screen understanding and browser interaction.
• Study and improve the agent's decision-making and learning loop.

Skills

Realtime systems

LLM agents

Browser/UI automation

Multimodal ML

Tools

Orchestration frameworks

Job description

Sable is assembling a world-class team to build Aidan, the first AI employee capable of leading customer calls across voice, vision and browser use. You will empower Aidan to orchestrate real-time modalities—voice, hands, eyes, and the Brain—while refining the verification loop to ensure robust performance in live scenarios.

You will work in a small, autonomous team, delivering end-to-end engineering across perception, action planning, and continuous learning.