Member of Technical Staff (Software Engineer, Agent Harness)
Tech Stack
Tag name is followed by "@" symbol and proficiency level value.
About proficiency levels:
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
AI @ 3
AWS @ 1
Kubernetes @ 1
Observability @ 3
Python @ 5
Rust @ 1
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
Details
Perplexity is seeking a software engineer to evolve the core agent harness that powers its flagship answer experience. The role focuses on building reliable, large-scale agent systems that accumulate knowledge and answer challenging questions.
The engineer will combine strong product judgment with single- and multi-agent engineering paradigms. Responsibilities include translating user needs and quantitative insights into improvements to agent orchestration, context management, performance, and reliability. The role also involves developing observability, rigorous evaluation, and developer experiences that make failures easy to reproduce and fix autonomously through methods such as autoresearch.
The agent harness will also support the training and evaluation of new models, creating a feedback loop between production experiences and model improvement.
Responsibilities
- Build and ship large-scale AI systems and services with high reliability, availability, and performance.
- Improve agent orchestration, context management, performance, and reliability.
- Apply single- and multi-agent engineering paradigms.
- Translate user needs and quantitative insights into product and system improvements.
- Develop excellent observability and rigorous evaluation processes.
- Build developer experiences that make failures easy to reproduce and fix autonomously.
- Support the training and evaluation of new models through the agent harness.
- Work across product and infrastructure advancements.
Requirements
- 2+ years of experience building and shipping large-scale AI systems and services with high reliability, availability, and performance.
- Strong software engineering skills, with particular attention to developer experience and future-proof system design.
- Proficiency with Python.
- Familiarity with Rust and cloud infrastructure such as AWS and Kubernetes is a bonus.
- BS, MS, or PhD in Computer Science, Engineering, or a related field, or equivalent experience.
Benefits
Full-time U.S. employees receive benefits including equity, health, dental, vision, retirement, fitness, commuter and dependent care accounts, and more. USD salary ranges apply only to U.S.-based positions. Final offer amounts depend on factors including experience and expertise.