Staff Software Engineer, AI Training Infrastructure

Engg; San Francisco (CA); On-site; USD 220,000 - 405,000; Full time

Job summary

Perplexity in San Francisco is seeking a Member of Technical Staff (Software Engineer, Training Infrastructure) to join the Training Infrastructure team. You will help build distributed systems that orchestrate AI workloads across global compute resources and define interfaces for experiments.

A strong candidate will have 2+ years deploying and shipping distributed systems, and experience with Python, Rust, C++, or Go, plus AWS and Kubernetes.

Qualifications

• 2+ years of experience developing and shipping distributed systems for compute-intensive workloads, with focus on reliability, scalability, and performance.
• Familiarity with RL, agent evaluation, or ML infrastructure, and interest in accelerating experimentation and model development.
• BS, MS, or PhD in Computer Science, Engineering, or related field, or equivalent practical experience.

Responsibilities

• Build distributed systems that orchestrate AI workloads across large-scale compute infrastructure.
• Define and implement interfaces for designing, managing, and launching experiments.
• Create tooling to make experiment results legible to humans and agents for retrospectives and closed-loop research.

Skills

Distributed systems

Backend languages (Python/Rust/C++/Go)

Cloud & Kubernetes

RL/ML infrastructure

Education

BS/MS/PhD in CS/Engineering

Tools

Python

Rust

C++

Go

AWS

Kubernetes

Job description

Perplexity in San Francisco is seeking a Member of Technical Staff (Software Engineer, Training Infrastructure) to join the Training Infrastructure team. You will help build distributed systems that orchestrate AI workloads across global compute resources and define interfaces for experiments.

A strong candidate will have 2+ years deploying and shipping distributed systems, and experience with Python, Rust, C++, or Go, plus AWS and Kubernetes.