DirectJobHub
Jobs · Engineering · California

Sr. Agentic AI Solutions Engineer

AMD · Santa Clara, CA · 3 wk ago
HybridEngineeringFull-time
We need to output HTML fragment with sections: h3 for headings like About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. Only include sections where posting has real specific content. We must not repeat job title as heading. No Job ID etc. Remove boilerplate like EEO, legal, etc. Also remove application instructions. We need to preserve all substantive info. We need to parse posting: It starts with "ADVANCE YOUR CAREER..." etc. Then "The Role" description. Then "The Person". Then "Key Responsibilities". Then "Preferred Experience". Then "Academic Credentials". Then "ALTERNATE LOCATION: Austin, TX". Then "This role is not eligible for visa sponsorship." Then "Benefits offered are described: AMD benefits at a glance." Then other boilerplate about unsolicited resumes, EEO, AI screening, etc. We need to remove boilerplate. We need sections: About the role (maybe from "The Role" paragraph). Responsibilities (Key Responsibilities). Requirements/Qualifications/Skills: we have "The Person" describing qualifications, "Preferred Experience" describing skills/experience, "Academic Credentials". Could combine into Requirements and Qualifications? The instruction: use headings for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). We can decide which sections have content. We have no explicit Pay or Schedule info. So we should not include Pay or Schedule sections. Benefits: posting says "Benefits offered are described: AMD benefits at a glance." That's vague, not specific. According to rule: never create Benefits section unless posting actually states them. It states "Benefits offered are described: AMD benefits at a glance." That's not specific detail; it's a reference. Likely we should not include Benefits because it's not substantive detail. The rule: "never create Pay/Schedule/Benefits sections unless the posting actually states them." It does state benefits are described elsewhere, but not actual benefits. I'd interpret that as not enough to create a Benefits section. So omit Benefits. Thus sections: About the role, Responsibilities, Requirements (maybe combine Person and Preferred Experience and Academic Credentials). Could also have Qualifications and Skills separate. But we have content for qualifications (The Person) and skills/experience (Preferred Experience) and academic credentials. Could split into Qualifications (The Person) and Skills (Preferred Experience) and maybe Education (Academic Credentials). But instruction: prefer a few substantive sections over many thin ones. So maybe combine into one "Requirements" section covering qualifications, experience, education. We'll produce:

About the role

... paragraph from The Role ...

Responsibilities

  • ... each responsibility ...

Requirements

  • ... combine Person, Preferred Experience, Academic Credentials ...
We must preserve all substantive info, not invent or summarize away detail. So we need to include each bullet point from Key Responsibilities as
  • . Also include each bullet from Preferred Experience and The Person and Academic Credentials as
  • under Requirements. We must not include boilerplate like "This is a deeply hands-on..." etc? That is part of The Role description; we should keep it. We need to extract The Role paragraph: It starts after "The Role" heading. Let's copy: "AMD's Software and Solutions Team is seeking a Senior Agentic AI Solutions Engineer to design, build, and deliver next-generation Agentic AI solutions built on AMD's heterogeneous computing portfolio. Working under the technical direction of AMD's Agentic AI Systems Architecture team, this role drives the hands-on development, integration, and performance optimization of end-to-end AI workflows that leverage AMD CPUs, GPUs, DPUs, networking, and software platforms to deliver secure, scalable, deterministic, and high-performance AI solutions for AMD's most strategic customers. Working across AMD's AI software ecosystem—including AMD + AMD (A+A) and AMD + NVIDIA (A+N) environments—and leveraging ROCm™, AMD Zen Software Studio, AMD Enterprise AI, and AMD's open-source software initiatives, this engineer will translate architectural vision into working customer solutions, reference implementations, and early customer proof-of-concepts (PoCs) across enterprise, cloud, sovereign AI, and edge deployments. Delivered solutions will focus on execution, performance, efficiency, data integrity, and security, including governance, compliance, and AI sovereignty, while providing implementation feedback that helps shape AMD silicon and software roadmaps. This is a deeply hands-on, customer-facing senior individual contributor role focused on solution delivery, technical implementation, and performance engineering. It is not a people management, program management, or alliance management position." That's the role description. Now The Person paragraph: "A deep hands-on engineer with strong expertise spanning AI infrastructure, virtualization systems, distributed systems, heterogeneous computing, intelligent data architectures, operating systems, cloud and Neo Cloud environments, networking, security architecture (including secure computing and cryptography), and large-scale AI platform design, complemented by broad experience across modern AI software stacks, enterprise solutions, and software engineering methodologies. This individual has a demonstrated history of delivering complex customer solutions and shipping working systems that span hardware, software, networking, security, and AI frameworks. The ideal candidate understands how frontier foundation models, agentic reasoning systems, orchestration frameworks, intelligent data structures, and heterogeneous compute resources interact to deliver scalable, deterministic, trustworthy, and production-ready AI systems. They thrive in customer-embedded environments where they can turn architectural vision into working code, benchmarks, and customer proof-points leading through technical excellence, deep coding ability, and solution delivery alongside the Fellow-level architect." We'll break into bullet points? Could keep as prose

    . But we need to use

    • for any list of duties/requirements/benefits. The Person is not a list but a paragraph of qualifications. Could be

      . The instruction: Use

      • for any list of duties/requirements/benefits, and

        for prose. So we can keep The Person as

        . Similarly Preferred Experience could be list? It's a series of bullet-like sentences but not explicitly list. Could be

        as well. However we have Key Responsibilities which are clearly a list (each sentence starts with a verb). We'll make those

        • . For Requirements we could combine Person, Preferred Experience, Academic Credentials into a

          or multiple

          . But we need to preserve all substantive info; we can keep them as paragraphs. But we also have Academic Credentials: "Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, Artificial Intelligence, Data Science, or related field preferred; advanced degree desired." That's a sentence. Thus Requirements section could contain multiple paragraphs: The Person, Preferred Experience, Academic Credentials. We'll output:

          About the role

          ...

          Responsibilities

          • Deliver on AMD's Agentic AI technical strategy across enterprise, cloud, sovereign AI, edge, and emerging AI deployment environments, translating architectural vision into working customer solutions, reference implementations, and PoCs.
          • Build and optimize end-to-end reference solutions that maximize heterogeneous execution across AMD CPUs, GPUs, DPUs, networking, memory, storage, and future accelerator technologies — supporting both AMD + AMD (A+A) and AMD + NVIDIA (A+N) deployment models.
          • Deliver intelligent multi-agent systems and orchestration frameworks that optimize execution across heterogeneous compute resources while balancing performance, latency, throughput, power efficiency, scalability, and cost.
          • Optimize Agentic AI execution end-to-end — task decomposition, multi-agent collaboration, workflow scheduling, reasoning pipelines, context management, token and inference efficiency, and resource allocation — balancing performance, cost, energy, security, and quality of service across heterogeneous AI infrastructure.
          • Implement innovative approaches for data optimization and intelligent data structures that improve retrieval efficiency, knowledge representation, inference performance, memory hierarchy utilization, and overall AI system scalability.
          • Deliver solutions that improve deterministic AI behavior, reduce hallucinations, and increase reliability, explainability, and trustworthiness of Agentic AI systems through grounding techniques, orchestration strategies, policy enforcement, validation frameworks, and intelligent workflow design.
          • Build trusted, secure AI solutions incorporating data provenance, data attestation, lineage tracking, governance, compliance, and AI sovereignty, leveraging confidential computing, trusted execution environments, hardware roots of trust, attestation, and cryptographic techniques.
          • Integrate frontier foundation models with retrieval systems, enterprise applications, orchestration frameworks, edge computing, sovereign AI environments, and hybrid cloud deployments.
          • Partner directly with strategic customers, hyperscalers, OEMs, ISVs, and open-source communities to embed, deliver, and validate AMD Agentic AI solutions in production-oriented environments, accelerating customer adoption and reference outcomes.
          • Partner with silicon architecture, software engineering, AI framework, and product management teams to provide implementation- and customer-driven feedback that informs future AMD silicon, software, and platform roadmaps.

          Requirements

          A deep hands-on engineer with strong expertise spanning AI infrastructure, virtualization systems, distributed systems, heterogeneous computing, intelligent data architectures, operating systems, cloud and Neo Cloud environments, networking, security architecture (including secure computing and cryptography), and large-scale AI platform design, complemented by broad experience across modern AI software stacks, enterprise solutions, and software engineering methodologies. This individual has a demonstrated history of delivering complex customer solutions and shipping working systems that span hardware, software, networking, security, and AI frameworks. The ideal candidate understands how frontier foundation models, agentic reasoning systems, orchestration frameworks, intelligent data structures, and heterogeneous compute resources interact to deliver scalable, deterministic, trustworthy, and production-ready AI systems. They thrive in customer-embedded environments where they can turn architectural vision into working code, benchmarks, and customer proof-points leading through technical excellence, deep coding ability, and solution delivery alongside the Fellow-level architect.

          Strong hands-on experience delivering large-scale AI infrastructure, distributed AI systems, cloud-native AI platforms, or enterprise AI solutions across heterogeneous computing architectures spanning CPUs, GPUs, DPUs, networking, storage, memory subsystems, virtualization, and cloud infrastructure. Strong understanding of modern AI software ecosystems, including ROCm™, AMD Enterprise AI, AMD Zen Software Studio, frontier foundation models, LLM inference, Retrieval-Augmented Generation (RAG), vector databases, model serving, and AI orchestration frameworks. Hands-on experience delivering Agentic AI solutions — multi-agent collaboration, orchestration frameworks, autonomous reasoning, planning, workflow optimization, and intelligent task execution. Deep hands-on expertise in data optimization, intelligent data structures, knowledge representation, vector databases, and retrieval architectures supporting enterprise-scale AI systems. Strong expertise in AI performance engineering — latency optimization, throughput, token efficiency, inference scalability, workload scheduling, and heterogeneous resource utilization. Experience delivering methodologies that improve AI reliability through deterministic execution, hallucination mitigation, grounding, explainability, and trusted AI outcomes; working knowledge of confidential computing, trusted execution environments, attestation, data provenance, and enterprise AI security. Deep hands-on coding ability with strong programming skills in Python, C, C++ and comfort in performance-critical code paths, profiling, and optimization. Background as a solutions engineer, applications engineer, or technical consultant at a major systems or platform company (e.g., HP, Dell, or comparable) or a top-tier consulting/integration organization is a strong plus. Excellent written and verbal communication; comfort embedding with customer engineering teams and presenting to senior customer stakeholders.

          Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, Artificial Intelligence, Data Science, or related field preferred; advanced degree desired.

          We should also include the ALTERNATE LOCATION: Austin, TX and visa sponsorship note? The location is substantive; we could include it in About the role or a separate section? There's no heading for Location in the list, but we could incorporate into About the role paragraph. The instruction: do not put Job ID/

    • Similar jobs