Distributed Systems Engineer — AI Training & Experiments

Perplexity; San Francisco (CA); In loco; USD 160.000 - 230.000; Tempo pieno

Descrizione del lavoro

Perplexity in San Francisco is seeking software engineers for the Training Infrastructure team, building distributed systems that orchestrate AI workloads across global compute infrastructure. You will design interfaces for experiments and create tools to make results legible to humans and agents.

Candidates should have 2+ years shipping distributed systems and proficiency in Python, Rust, C++, or Go, plus cloud and Kubernetes.

Competenze

• ,"BS, MS, or PhD in Computer Science, Engineering, or a related field, or equivalent practical experience."
• "2+ years of experience developing and shipping distributed systems for compute-intensive workloads"
• "Proficiency in a general-purpose backend language such as Python, Rust, C++, or Go"
• "Familiarity with RL, agent evaluation, or ML infrastructure"

Mansioni

• Build distributed systems that orchestrate AI workloads across global compute infrastructure.
• Define novel user and agent interfaces for designing, managing, and launching experiments.
• Create tools that make experiment results legible to humans and agents alike, enabling retrospectives and autoresearch.

Conoscenze

Distributed systems

Backend development

Python

Rust

C++

Go

AWS

Kubernetes

Reinforcement learning

ML infrastructure

Formazione

BS/MS/PhD in CS/Engineering or related field

Strumenti

AWS

Kubernetes

Descrizione del lavoro

Perplexity in San Francisco is seeking software engineers for the Training Infrastructure team, building distributed systems that orchestrate AI workloads across global compute infrastructure. You will design interfaces for experiments and create tools to make results legible to humans and agents.

Candidates should have 2+ years shipping distributed systems and proficiency in Python, Rust, C++, or Go, plus cloud and Kubernetes.