Principal AI Agent Capabilities Engineer

Precision Labs; San Francisco, Northern (CA, KY); Hybrid; USD 180,000 - 280,000; Full time

Job summary

Perplexity Computer is building agent capabilities that transform knowledge into action. You will evaluate frontier models, design scalable agent systems, and own the lifecycle from prototyping to production launch, monitoring, and iteration.

Join a team that shapes how users and agents solve complex tasks with secure, observable architectures and a strong emphasis on craftsmanship, ownership, and rapid delivery.

Qualifications

• Typically 6+ years of professional software engineering experience.
• Strong software engineering fundamentals with AI/ML products or distributed systems.
• Experience owning the AI product lifecycle, data analysis, evaluation, and production monitoring.
• Practical experience in agent harnesses, tool use, context engineering, model evaluation, browser automation, or long-running task execution.
• Strong product judgment and ability to ship durable AI/ML solutions with measurable impact.
• Genuine interest in frontier AI capabilities and agent systems.

Responsibilities

• > Evaluate frontier models against real user tasks and turn advances into production agent systems.
• > Improve agents’ planning, tool use, context handling, error recovery, and long-running task reliability.
• > Apply state-of-the-art ML/LLM techniques to design scalable agent capabilities and architectures.
• > Own agent behavior end-to-end, from user interfaces to backend services and monitoring.
• > Build secure, observable agent systems with tracing and safeguards.
• > Collaborate with PM, Data Science, and Research to identify high-impact opportunities.

Skills

Software engineering

AI/ML products

Agent harnesses

Tool use

Context engineering

Model evaluation

Browser automation

Long-running tasks

Tools

Python

Go

Rust

PostgreSQL

DynamoDB

AWS

TypeScript

Job description

Perplexity Computer is building agent capabilities that transform knowledge into action. You will evaluate frontier models, design scalable agent systems, and own the lifecycle from prototyping to production launch, monitoring, and iteration.

Join a team that shapes how users and agents solve complex tasks with secure, observable architectures and a strong emphasis on craftsmanship, ownership, and rapid delivery.