Define architecture for hierarchical AI control loops, including
Perception (sensor fusion, VLMs, state estimation)
World modeling and scene understanding
Task and motion planning
Low-latency control and actuation
Learning and adaptation loops
Establish timing models and system-level constraints
Deterministic execution for control loops (µs–ms)
Bounded latency for perception/planning (ms-scale)
Guide architectural decisions to
Minimize latency, jitter, and data movement
Optimize compute utilization across heterogeneous systems
Influence compute-software co-design across CPU, GPU, and accelerators
Lighthouse Customer Co‑Engineering
Act as architectural lead for strategic/lighthouse customers, guiding
System design decisions
Performance trade-offs
Deployment architecture
Influence customer implementations to align with
Platform best practices
Scalable architectural patterns
Translate real-world constraints (latency, power, safety) into
Platform requirements
Architecture refinements
Lead deep technical engagements, including
Architecture and design reviews
Performance tuning strategies
System-level debugging approaches
Influencing Robotics Reference Architectures, Platform Architecture & SDKs
Define reference architecture for complex robotic systems (humanoids, high-DoF manipulators, mobile manipulation platforms), establishing industry-leading blueprints for Physical AI systems
Influence architectural partitioning strategies across
On-robot compute (real-time loops)
Edge/accelerator subsystems
Cloud (training, simulation, fleet learning)
Provide architectural guidance on
Whole-body control integration
Locomotion and balance systems
Dexterous manipulation pipelines
Multi-modal perception stacks
Serve as a bridge between lighthouse deployments and platform evolution, translating system-level insights into:
SDK feature direction
Runtime and middleware enhancements
Reference pipeline abstractions
Shape the roadmap of
Physical AI SDK and runtime frameworks
Robotics middleware integrations (ROS2 and beyond)
Dataflow and scheduling architectures for deterministic execution
Identify systemic gaps and influence solutions in
Scheduling and orchestration models
Memory and dataflow efficiency
Inter-process/inter-node communication
Real-time guarantees and QoS mechanisms
Drive the creation of scalable architectural patterns, including
Reusable operator graphs and pipelines
Standardized deployment topologies
Benchmark and validation frameworks
Ensure that lighthouse customer learnings are abstracted and generalized into:
Repeatable reference architectures
Platform capabilities consumable across a wide customer base
PREFERRED EXPERIENCE
Experience in Robotics, Autonomous systems AI architecture, definition and development, RL work, Sim-to-real, cloud-to-sim, real-to-sim/cloud from AI perspective.
Proven technical leadership experience influencing external and internal stakeholders
Deep understanding of
AI inference runtimes and deployment tradeoffs
System architecture level CPU/GPU/NPU scheduling and contention
System‑level performance, latency, and isolation
Software frameworks and usage (multimedia, ROS2, OpenCV, gstreamer etc.)
Industry leading SW inference frameworks (vLLM etc.), runtimes, tools
Performance bottleneck, characterization
Determinism, real-time and safety considerations in mixed-criticality systems
Ability to engage credibly with customer’s engineering leaders, AI architects
Track record of transforming customer deployments into platform and roadmap feedback
Hands-on architects who can guide engineers, debug problems, create innovative PoCs as well as abstract unnecessary details for executive presentations
Familiarity/experience with AMD GPU and NPU AI SW stacks and tools will be a plus
ACADEMIC CREDENTIALS
Bachelor’s or Master’s in Electrical Engineer, Computer Engineering, Computer Science, or a closely related field
This role is not eligible for visa sponsorship.
#LI-JG1
Benefits 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.
THE ROLE
We are seeking a Robotics AI Architect to define and scale next-generation Physical AI systems, with a focus on complex robotic platforms (including humanoids). This role emphasizes architectural leadership across multi-layer AI control loops and tight collaboration with lighthouse customers to achieve production-grade performance targets.
THE PERSON
As a key technical authority, you will synthesize learnings from real-world deployments and translate them into platform-defining capabilities, shaping the roadmap for our AI SDKs, runtime, and reference architectures to enable broad ecosystem scalability.
KEY RESPONSIBILITIES
End-to-End AI Control Loop Architecture (Core Focus)
Define architecture for hierarchical AI control loops, including
Perception (sensor fusion, VLMs, state estimation)
World modeling and scene understanding
Task and motion planning
Low-latency control and actuation
Learning and adaptation loops
Establish timing models and system-level constraints
Deterministic execution for control loops (µs–ms)
Bounded latency for perception/planning (ms-scale)
Guide architectural decisions to
Minimize latency, jitter, and data movement
Optimize compute utilization across heterogeneous systems
Influence compute-software co-design across CPU, GPU, and accelerators
Lighthouse Customer Co‑Engineering
Act as architectural lead for strategic/lighthouse customers, guiding
System design decisions
Performance trade-offs
Deployment architecture
Influence customer implementations to align with
Platform best practices
Scalable architectural patterns
Translate real-world constraints (latency, power, safety) into
Platform requirements
Architecture refinements
Lead deep technical engagements, including
Architecture and design reviews
Performance tuning strategies
System-level debugging approaches
Influencing Robotics Reference Architectures, Platform Architecture & SDKs
Define reference architecture for complex robotic systems (humanoids, high-DoF manipulators, mobile manipulation platforms), establishing industry-leading blueprints for Physical AI systems
Influence architectural partitioning strategies across
On-robot compute (real-time loops)
Edge/accelerator subsystems
Cloud (training, simulation, fleet learning)
Provide architectural guidance on
Whole-body control integration
Locomotion and balance systems
Dexterous manipulation pipelines
Multi-modal perception stacks
Serve as a bridge between lighthouse deployments and platform evolution, translating system-level insights into:
SDK feature direction
Runtime and middleware enhancements
Reference pipeline abstractions
Shape the roadmap of
Physical AI SDK and runtime frameworks
Robotics middleware integrations (ROS2 and beyond)
Dataflow and scheduling architectures for deterministic execution
Identify systemic gaps and influence solutions in
Scheduling and orchestration models
Memory and dataflow efficiency
Inter-process/inter-node communication
Real-time guarantees and QoS mechanisms
Drive the creation of scalable architectural patterns, including
Reusable operator graphs and pipelines
Standardized deployment topologies
Benchmark and validation frameworks
Ensure that lighthouse customer learnings are abstracted and generalized into:
Repeatable reference architectures
Platform capabilities consumable across a wide customer base
PREFERRED EXPERIENCE
Experience in Robotics, Autonomous systems AI architecture, definition and development, RL work, Sim-to-real, cloud-to-sim, real-to-sim/cloud from AI perspective.
Proven technical leadership experience influencing external and internal stakeholders
Deep understanding of
AI inference runtimes and deployment tradeoffs
System architecture level CPU/GPU/NPU scheduling and contention
System‑level performance, latency, and isolation
Software frameworks and usage (multimedia, ROS2, OpenCV, gstreamer etc.)
Industry leading SW inference frameworks (vLLM etc.), runtimes, tools
Performance bottleneck, characterization
Determinism, real-time and safety considerations in mixed-criticality systems
Ability to engage credibly with customer’s engineering leaders, AI architects
Track record of transforming customer deployments into platform and roadmap feedback
Hands-on architects who can guide engineers, debug problems, create innovative PoCs as well as abstract unnecessary details for executive presentations
Familiarity/experience with AMD GPU and NPU AI SW stacks and tools will be a plus
ACADEMIC CREDENTIALS
Bachelor’s or Master’s in Electrical Engineer, Computer Engineering, Computer Science, or a closely related field
This role is not eligible for visa sponsorship.
#LI-JG1
Benefits 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.