GenAI Software Development Architect (Santa Clara)

WHAT YOU DO AT AMD CHANGES EVERYTHING At AMD, our mission is to build great products that accelerate next-generation computing experiences—from AI and data centers, to PCs, gaming and embedded systems. Grounded in a culture of innovation and collaboration, we believe real progress comes from bold ideas, human ingenuity and a shared passion to create something extraordinary. When you join AMD, you’ll discover the real differentiator is our culture. We push the limits of innovation to solve the world’s most important challenges—striving for execution excellence, while being direct, humble, collaborative, and inclusive of diverse perspectives. Join us as we shape the future of AI and beyond. Together, we advance your career. THE ROLE: We are building an AI-native hardware and firmware validation platform from the ground up — one where LLMs, RAG pipelines, autonomous agents, and knowledge graphs are the core of how the system works, not an add-on. As the Software Development Architect, you will own the end-to-end technical design of this platform: multi-agent orchestration, retrieval-augmented knowledge systems, MCP server infrastructure, and the engineering standards that make all of it reliable at scale. This role sits within the Global Cluster Engineering organization, where you will develop software that powers distributed infrastructure at global scale. You will work closely with validation engineers, hardware teams, and leadership to translate domain requirements into a production-grade AI-native system. This is a hands-on role — you will write code, drive technology decisions, and directly mentor engineers.THE PERSON: Experience: software development experience, with at least 4 years in architecture, staff, or principal engineer roleAI-Native Systems: Deep, hands-on experience designing and shipping production AI-native systems — not just LLM API integration, but the full stack: RAG pipelines, agent orchestration, tool use, multi-agent coordination, and LLM evaluationLLM Fundamentals: Strong understanding of how LLMs work in practice — context windows, grounding, hallucination failure modes, prompt engineering, model selection, and how behavior changes across providers and versionsRetrieval Systems: Proven experience with vector search, embedding models, hybrid retrieval, reranking pipelines, and knowledge graph-augmented RAGCore Skills: Strong proficiency in one or more modern programming languages such as Python, TypeScript/Node.js, Go, Java, C#, or Rust, with demonstrated ability to build and operate production-scale services. Python experience is preferred due to the AI/ML ecosystemEngineering excellence: Async programming, API design, distributed systems, clean code practices. Experience designing for reliability in automated/unattended environments — crash recovery, audit trails, state management, observability. Strong written communication — architecture docs, design specs, and engineering standards that outlast your tenure. Track record of setting engineering standards that teams followHardware Affinity: Experience working closely with hardware teams — servers, networking equipment, or compute infrastructure — with an understanding of how software interacts with physical systemsCloud Infrastructure: Experience with AWS, Azure, or GCP — infrastructure provisioning, managed services, networking, and deploying production workloads at scaleAI Tooling: Demonstrated use of AI coding assistants and LLM-powered developer tools (Claude Code, GitHub Copilot, Cursor, etc.) to accelerate design, development, and documentationKEY RESPONSIBILITIES: Platform Architecture: Design and own the architecture of an AI-native validation platform where autonomous LLM agents plan, execute, and analyze hardware and firmware test campaigns end-to-end — without a human in the loopRAG System Design: Architect the full retrieval-augmented generation stack — document ingestion pipelines, chunking strategies, embedding models, vector stores, knowledge graph backends, hybrid search, cross-encoder and LLM-based reranking — ensuring agents have accurate, grounded knowledge at query timeAgent Orchestration: Define multi-agent dispatch patterns, context window management strategies, anti-hallucination contracts, tool-use boundaries, inter-agent communication protocols, and crash recovery mechanisms for long-running unattended runsMCP Integration: Own the integration architecture between the agent layer (Claude Code / Model Context Protocol), the knowledge backend (Qdrant, Neo4j / LightRAG), and external systems (Slack, Jira, Confluence, GitHub) or equivalentEngineering Standards: Establish and enforce AI-native development standards — prompt design, skill authoring, agent contract specifications, artifact schemas, and evaluation methodology for LLM outputsLLM Reliability: Lead the team's approach to building trustworthy agentic systems — fabrication detection, context compaction recovery, output validation, and post-run audit infrastructureTechnology Evaluation: Continuously evaluate new LLM capabilities, model releases, embedding models, and agentic frameworks; make pragmatic adoption decisionsAI Services: Design and implement scalable, low-latency AI services powering metadata generation, feature extraction, and knowledge retrieval across the validation platformAgentic AI Deployment: Develop and deploy agentic AI solutions — autonomous agents, multi-agent orchestration frameworks, and LLM-powered workflows — that transform hardware validation, firmware QA, and lab operationsStakeholder Collaboration: Partner with engineering peers, validation engineers, and business stakeholders to understand requirements and translate them into flexible, future-proof design solutionsSecurity & Compliance: Ensure AI/ML systems comply with security standards and best practices, addressing data privacy and protection concerns across all LLM integrations and knowledge pipelinesEnd-to-End Ownership: Own the platform end-to-end — from project estimation and architecture review through coding, deployment, and post-launch measurementOperational Excellence: Build resilient systems with strong observability; establish automated testing, monitoring, and CI/CD pipelines using infrastructure-as-code tools (Terraform); lead root-cause analysis and drive continuous reliability improvementsTeam Leadership: Mentor software developers, conduct design reviews, and set the technical bar for the teamPREFERRED EXPERIENCE: Background in the semiconductor or datacenter industry — hardware validation, firmware development, or silicon bring-upExperience with network hardware (NICs, switches, GPUs) or associated diagnostics (PCIe, RDMA, etc)Familiarity with the Model Context Protocol (MCP) or agentic platforms (LangGraph, CrewAI, AutoGen)Published work, open source contributions, or talks in the AI/LLM spaceData Engineering & Analytics: Experience with data pipeline design, ETL workflows, data warehousing, or analytics platforms is a plusACADEMIC CREDENTIALS: BS or MS Degree in Computer Science, Electrical Engineering, or related fieldLOCATION: Santa Clara Austin or Seattle or Secaucus#LI-KW1Benefits offered are described: AMD benefits at a glance.AMD does not accept unsolicited resumes from headhunters, recruitment agencies, or fee-based recruitment services. AMD and its subsidiaries are equal opportunity, inclusive employers and will consider all applicants without regard to age, ancestry, color, marital status, medical condition, mental or physical disability, national origin, race, religion, political and/or third-party affiliation, sex, pregnancy, sexual orientation, gender identity, military or veteran status, or any other characteristic protected by law. We encourage applications from all qualified candidates and will accommodate applicants’ needs under the respective laws throughout all stages of the recruitment and selection process.AMD may use Artificial Intelligence to help screen, assess or select applicants for this position. AMD’s “Responsible AI Policy” is available here.This posting is for an existing vacancy.