Member of Technical Staff (Software Engineer, AI Systems)

Perplexity • San Francisco, CA

6d • $220,000 - $405,000

About The Position

Perplexity is looking for a software engineer to evolve the core agent harness that powers our flagship answer experience. You will combine strong product judgment with both single- and multi-agent engineering paradigms to help us build the world’s foremost agent for accumulating knowledge and answering our society’s most challenging questions. You will translate user needs and quantitative insights into improvements to the agent’s orchestration, context management, performance, and reliability. The ideal candidate understands that reliable agents at global scale require excellent observability, rigorous evaluation, and a developer experience that makes failures easy to reproduce and fix autonomously through methods like autoresearch. The same harness will power the training and evaluation of new models, creating a tight feedback loop between production experiences and model improvement that is central to Perplexity’s mission.

Requirements

• 2+ years of experience building and shipping large-scale AI systems and services with high reliability, availability, and performance.
• Strong software engineering craft, with special care for developer experience and designing future-proof systems that can evolve along with a rapidly advancing AI landscape.
• Proficiency with Python.
• BS, MS, or PhD in Computer Science, Engineering, or related fields (or equivalent experience).

Nice To Haves

• Familiarity with Rust.
• Familiarity with cloud infrastructure (AWS, Kubernetes).

Responsibilities

• Build products and systems that define how AI empowers humans to think, decide and act.
• Ship fast and work between both product and infrastructure advancements in our technology.
• Evolve the core agent harness that powers our flagship answer experience.
• Combine strong product judgment with both single- and multi-agent engineering paradigms to help us build the world’s foremost agent for accumulating knowledge and answering our society’s most challenging questions.
• Translate user needs and quantitative insights into improvements to the agent’s orchestration, context management, performance, and reliability.
• Ensure reliable agents at global scale require excellent observability, rigorous evaluation, and a developer experience that makes failures easy to reproduce and fix autonomously through methods like autoresearch.
• Power the training and evaluation of new models, creating a tight feedback loop between production experiences and model improvement.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior