Software Engineer, Applied AI Infrastructure
Core
Building a rigorous closed-loop evaluation system, agent platform, and autoresearch infrastructure to make AI agents trustworthy and autonomous within Nuro's engineering organization.
Role type
Senior IC Applied AI Infrastructure Engineer (Agent Systems & Evaluation)
Builds
Closed-loop measurement layers, agent runtime platforms, and automated research loops for LLM-based agents.
Domain
Autonomous driving / Physical AI / LLM Infrastructure
Deliverable
production ML models | infrastructure
Required skills
LLM training dynamics (SFT, RL, distillation), LLM inference optimization (KV-cache, speculative decoding), production agent system architecture, distributed systems at scale, Python, post-training evaluation, experiment orchestration, isolation/security patterns
Preferred skills
Go/C++/Rust, developer experience platform engineering, open tool-integration protocols, multi-agent coordination
Technologies
Python, LLM frameworks, cloud infrastructure, queuing systems, observability tools
Responsibilities
Build closed-loop measurement layers to track agent acceptance/revert/override rates; take autoresearch loops from assisted to unattended; design isolation and permissioning models for agents acting on production systems; build agent-powered tooling for code generation, review, and debugging.
Seniority
Senior, hands-on IC
