Staff Data Engineer: Agentic AI & Production Pipelines

AMD; Indiana (PA); On-site; USD 150,000 - 230,000; Full time

Job summary

AMD is seeking a data engineer with applied AI focus to build agentic AI systems on AMD Instinct GPU infrastructure. You will design memory and context layers, develop an engaging memory and data pipelines, and maintain a library of agent-enabled capabilities.

You will work on evaluation and hardening against real-world failures while collaborating with customer teams. Your role spans external deployments and internal intelligence layers, requiring strong Python, data engineering, and

Qualifications

• Bachelor’s or Master’s degree in Computer Science, Computer Engineering, or equivalent practical experience.
• Strong production data engineering background with scalable pipelines and data quality.

Responsibilities

• Build production agentic AI systems on AMD Instinct GPU infrastructure: agent orchestration, tool calling, skills frameworks, and streaming inference integration.
• Design and operate memory and context data layer for agentic apps: vector/graph/relational stores, embedding pipelines, retrieval and provenance policies.
• Develop engagement memory and fleet data infrastructure: telemetry ingestion, incident histories, and field knowledge into structured forms.
• Maintain skills library: versioned, agent-executable encodings with testing and review gates.
• Build evaluation infrastructure: regression suites, LLM-as-judge, behavioral test harnesses, production monitoring.
• Harden systems against real-world failure modes: prompt injection, data poisoning, tool-misuse.
• Create reference architectures and open artifacts for AMD agentic workloads and upstream communities.
• Partner with customer-facing engineers on live engagements: deployments in customer environments.

Skills

Python

Kubernetes

Data pipelines

Agentic AI

Memory design

Education

Bachelor or Master in CS/CE

Tools

ROCm

vLLM

SGLang

Job description

AMD is seeking a data engineer with applied AI focus to build agentic AI systems on AMD Instinct GPU infrastructure. You will design memory and context layers, develop an engaging memory and data pipelines, and maintain a library of agent-enabled capabilities.

You will work on evaluation and hardening against real-world failures while collaborating with customer teams. Your role spans external deployments and internal intelligence layers, requiring strong Python, data engineering, and