Senior Production Engineer, Tooling & Frameworks

CoreWeave•Sunnyvale, CA

1d•$139,000 - $204,000•Hybrid

About The Position

Production Engineering ensures CoreWeave’s cloud runs with world-class reliability, performance, and operational excellence. Herd is our newest innovation: an agentic AI platform that serves as CoreWeave’s intelligent SRE assistant - combining AI reasoning, data infrastructure, and observability into an autonomous operational intelligence layer for internal use. As a Production Engineer on Herd, you’ll define and build the systems that power a scalable agentic ecosystem. You’ll design distributed services and data pipelines that process, embed, and retrieve operational knowledge at scale, enabling LLM-powered agents to work alongside human engineers in production. This is a hands-on role at the intersection of AI operations, distributed systems, and data infrastructure.

Requirements

5+ years in software or infrastructure engineering building and operating distributed systems at scale.
Proficiency in Python (or similar), with experience delivering production microservices and data pipelines.
Expertise in Kubernetes, container orchestration, and cloud-native architectures.
Strong understanding of data systems (streaming, indexing, caching, ETL).
Demonstrated experience designing for scalability, fault tolerance, and performance.

Nice To Haves

Experience building RAG systems or embedding-based search.
Familiarity with vector databases and text/signal retrieval systems.
Experience developing or deploying agentic AI systems (LLM-based automation, AI observability).
Strong distributed-systems background (consensus, message buses, job orchestration, eventual consistency).
Experience designing data schemas and APIs for knowledge representation and operational reasoning.
Familiarity with ChatOps frameworks or workflow orchestration (Temporal, Argo, Airflow).
Background in MLOps, AI infrastructure, or platform reliability engineering.
Experience with observability frameworks (Prometheus, Grafana, OpenTelemetry) and using telemetry for automated reasoning or remediation.

Responsibilities

Architect and build large-scale distributed systems that power AI SRE Platforms.
Design data infrastructure for AI reasoning (embedding generation, context retrieval, vector stores) optimized for real-time operational queries.
Build agent orchestration and lifecycle components so agents can communicate, delegate, and reason collectively across CoreWeave systems.
Integrate AI SRE Platform with a large number of internal systems (Kubernetes, observability platforms, etc.) to enable end-to-end automation and insights.
Lead architectural design discussions and set technical direction for AI-driven reliability systems.
Partner across Production Engineering, Data Engineering, ML Infrastructure, and Platform to operate AI SRE as a high-availability platform embedded in critical reliability workflows.
Develop services that interpret telemetry, detect anomalies, and generate RCA (root cause analysis) and PIR (post-incident review) artifacts; trigger automated mitigations where appropriate.
Codify operational best practices into services, APIs, and Kubernetes-native components.
Participate in an on-call rotation supporting the systems you build.