AI Ops Engineer

Job Title: AI Ops Engineer
Experience: 3–5 years
About the Role
We are seeking a hands-on and proactive AI Ops Engineer to operationalize and
support the deployment of large language model (LLM) workflows, including agentic AI
applications, across Marvell’s enterprise ecosystem.
This role requires strong prompt engineering capabilities, the ability to triage AI pipeline
issues, and a deep understanding of how LLM-based agents interact with tools,
memory, and APIs. You will be expected to diagnose and remediate real-time problems,
from prompt quality issues to model behavior anomalies.

Key Responsibilities
Design, fine-tune, and manage prompts for various LLM use cases tailored to Marvell’s
enterprise operations.
Operate, monitor, and troubleshoot agentic AI applications, including identifying whether
issues stem from:
Prompt quality or structure
Model configuration or performance
Tool usage, API failures, or memory/recall issues
Build diagnostics and playbooks to triage LLM-driven failures, including handling fallback
strategies, retries, or re-routing to human workflows.
Collaborate with architects, ML engineers, and DevOps to optimize agent orchestration
across platforms like LangGraph, CrewAI, AutoGen, or similar.
Support integration of agentic systems with enterprise apps like Jira, ServiceNow, Glean, or
Confluence using REST APIs, webhooks, and adapters.
Implement observability and logging best practices for model outputs, latency, and agent
performance metrics.
Contribute to building self-healing mechanisms and alerting strategies for production-grade
AI workflows.
Required Qualifications

3–6 years of experience in software engineering, DevOps, or ML Ops with exposure to
AI/LLM workflows.
Strong foundation in prompt engineering and experience with LLMs like GPT, Claude,
LLaMA, etc.
Practical understanding of AIOps platforms or operational AI use cases (incident triage,
log summarization, root cause analysis, etc.).
Exposure to agentic AI architectures, such as LangGraph, AutoGen, CrewAI, etc.
Familiarity with scripting (Python), RESTful APIs, and basic system debugging.
Strong analytical skills and the ability to trace issues across multi-step pipelines and
asynchronous agents.

Good-To-Have
Glean
DevRev
Codium
Cursor
Atlassian AI
Databricks Mosaic AI