AI Engineer

CoreWeave; Sunnyvale (CA); On-site; USD 140,000 - 230,000; Full time

Job summary

CoreWeave is seeking an AI Engineer to design, build, and operate systems used during real customer incidents. You will balance AI and deterministic logic, own agent behavior, and ensure safeguards and data grounding.

You will deploy services on Kubernetes, contribute to CI/CD pipelines, and collaborate with support engineers during live investigations to optimize automation and reliability.

Qualifications

• 3+ years of professional software engineering experience.
• Proficiency in Go or Python.
• Experience with observability for LLM applications (tracing, cost, latency).
• Design controls for automation on production systems (approval steps, idempotency, audit trails).
• Hands-on experience with agent patterns: tool calls, multi-step orchestration, handling incorrect model output.

Responsibilities

• Design, build, and operate systems used during real customer incidents.
• Define balance between AI and deterministic logic and own agent behavior.
• Deploy and operate services on Kubernetes, with CI/CD pipelines.
• Work with support engineers and domain experts during live investigations.
• Ground model outputs in authoritative data and implement safeguards.

Skills

Go

Python

Observability

Kubernetes

CI/CD

Agent patterns

Retrieval systems

LLM tooling

Security & compliance

Job description

• The Field Engineering organization at CoreWeave supports the clients running some of the largest AI workloads in the world
• This team builds the tooling that engineers rely on to identify and diagnose customer issues faster
• We work where AI and deterministic based systems meet live production infrastructure, so what we build has to be accurate, has to show its reasoning, and has to earn the trust of engineers who are the real experts solving issues
• As an AI Engineer, you’ll design, build, and operate systems that engineers use during real customer incidents
• You will define the right balance between deterministic and AI systems, and work with both. For AI systems, you’ll own the agent logic, the retrieval behind it, the safeguards that stop it from acting on bad information, and the measurement that tells us whether it’s actually helping
• This is a full-ownership role: you ship your own services and stay responsible for how they behave in production
• You’ll work closely with the engineers who use what you build
• Design and build AI agents that investigate and act on real customer issues
• Build the safeguards that make that safe: grounding model output in authoritative data, requiring human approval before consequential actions, and defining the conditions under which the system should stop
• Build and evaluate retrieval over a large body of historical support data, measured against how experienced engineers handled the same problems
• Determine the optimal balance between AI and deterministic logic
• Instrument these systems for quality, cost, and adoption, then use what you learn to decide what to build next
• Deploy and operate your own services on Kubernetes, including supporting them when something breaks
• Partner with support engineers and domain experts, including sitting in on live investigations to see where automation helps and where it gets in the way

3+ years of professional software engineering experience, with strong proficiency in Go or PythonExperience with observability for LLM applications, including tracing, cost, and latencyExperience designing controls for automation that acts on production systems, such as approval steps, idempotency, and audit trailsHands-on experience with agent patterns: tool and function calling, multi-step orchestration, and handling incorrect model outputWorking knowledge of Kubernetes and experience deploying services through CI/CDExperience integrating with third party APIs, including authentication and secrets handlingExperience building and evaluating retrieval systems, including measuring quality against a baseline using held-out dataExperience diagnosing failures in LLM applications and fixing the underlying cause rather than the symptomExperience taking an AI application from prototype to production use and supporting it afterwardThis position requires access to export controlled information. To conform to U.S. Government export regulations applicable to that information, applicant must either be (A) a U.S. person, defined as a (i) U.S. citizen or national, (ii) U.S. lawful permanent resident (green card holder), (iii) refugee under 8 U.S.C. § 1157, or (iv) asylee under 8 U.S.C. § 1158, (B) eligible to access the export controlled information without a required export authorization, or (C) eligible and reasonably likely to obtain the required export authorization from the applicable U.S. government agencyExperience building conversational or chat-based interfaces for internal usersExperience testing both deterministic and non-deterministic systemsA background in infrastructure, SRE, or technical support, and comfort reading logs and metrics from a live systemOpen source contributions to AI tooling or agent frameworksExperience with fine-tuning or building training datasets from production data