FlexAI
NoC Designer R&D HW · San Francisco (Bay Area) ·
FlexAI, California, Missouri, United States, 65018
Join FlexAI:
FlexAI is at the forefront of revolutionizing AI computing by reengineering infrastructure at the system level. Our groundbreaking architecture, combined with sophisticated software intelligence, abstraction, and an orchestration layer, allows developers to leverage a diverse array of compute, resulting in efficient, more reliable computing at a fraction of the cost.The rapid evolution of machine intelligence has created a need for a new system architecture capable of handling high memory capacity and bandwidth. These are critical bottlenecks in pushing machine intelligence to the next level, where compute demand is expected to increase up to 1000 times current levels.FlexAI has pioneered a groundbreaking solution to tackle these memory challenges. Our innovative compute architecture ensures a well-balanced distribution of memory bandwidth, capacity, and compute density, ensuring maximum utilization of system resources. This architecture is the cornerstone of our datacenter-in-a-box concept, which is enabled by our universal AI compute cloud service. Our hardware solutions are built for seamless deployment with our own AI cloud offerings and other cloud service providers worldwide, setting new standards in performance and efficiency.We are looking for a
NoC Designer
who can architect and design the next generation of Network-on-Chip (NoC), a critical component of our groundbreaking system architecture.Position Overview:
As the
NoC Designer,
you will design and optimize the on-chip communication fabric that connects processing elements, memory hierarchies, and AI accelerators within a System-on-Chip (SoC). Your focus will be on developing scalable and high-performance NoC designs that meet AI-driven applications' bandwidth, latency, and power requirements. You will closely collaborate with SoC architects, AI hardware teams, and software engineers to ensure that the NoC supports the complex data flows required by AI workloads. This is your opportunity to lead, innovate, and usher in a new era of AI hardware design.Success at FlexAI requires an entrepreneurial spirit and startup mindset: the ability to rapidly iterate and make meaningful progress while staying focused on our mission to deliver more compute with less complexity.What you’ll do:NoC Architecture Design:Design and develop high-performance Network-on-Chip (NoC) architectures that interconnect processing cores, AI accelerators, memory subsystems, and peripheral IP blocks.Optimize the NoC for AI-specific workloads such as deep learning inference and training, ensuring that the interconnect can handle high-bandwidth data transfers with low latency.Dataflow and Bandwidth Optimization:Analyze and optimize dataflow patterns for AI workloads, ensuring that the NoC design minimizes congestion, latency, and power consumption while maximizing throughput.Implement advanced techniques such as packet-based routing, dynamic scheduling, and traffic prioritization to manage the data traffic within the SoC efficiently.Scalability and Configurability:Architect scalable NoC designs that support various SoC configurations, from edge AI devices to high-performance AI accelerators.Design modular and configurable NoC components that are easily integrated into different SoC architectures.Performance Modeling and Analysis:Develop performance models and simulation environments to evaluate the NoC architecture under various AI workload scenarios.Perform trade-off analysis between performance, power, and area (PPA) to optimize the NoC design for AI SoCs.Power and Area Efficiency:Work on power-saving techniques such as dynamic voltage and frequency scaling (DVFS), power gating, and clock gating to optimize the NoC for power efficiency.Collaborate with physical design teams to ensure that the NoC design meets area and layout constraints.Cross-Functional Collaboration:Work closely with SoC architects, AI hardware designers, and memory subsystem teams to ensure the NoC design meets system-level requirements.Collaborate with software and firmware teams to optimize communication protocols and ensure efficient data movement across the SoC.What you’ll need to be successful:Bachelor’s, Master’s, or Ph.D. in Electrical Engineering, Computer Engineering, or a related field.5+ years of experience in NoC architecture and design, with a focus on AI, high-performance computing (HPC), or similar workloads.Proven track record in designing and delivering NoC solutions for complex SoCs.Experience with industry-standard NoC IP, network topologies, and interconnect protocols (e.g., AMBA, AXI, CHI).Strong knowledge of NoC architectures, including routing algorithms, flow control, and quality of service (QoS) mechanisms.Experience in performance modeling, simulation, and analysis tools for NoC designs.Proficiency in hardware description languages (e.g., Verilog, VHDL) and NoC design tools.Experience with RTL design, synthesis, and timing closure for custom IP blocks.Strong knowledge of AI hardware architectures, including neural processing units (NPUs), tensor cores, and other AI-specific accelerators.Proficiency in microarchitecture design principles, pipeline optimization, and memory hierarchy design.Experience with performance modeling tools and hardware description languages (e.g., Verilog, VHDL).Familiarity with high-level synthesis (HLS) tools and techniques.Strong analytical and problem-solving skills with a focus on system-level optimization.Ability to work effectively in cross-functional teams, including hardware, software, and verification groups.Excellent communication skills for both technical and non-technical audiences.Preferred Skills:Experience with AI/ML software frameworks (e.g., TensorFlow, PyTorch) and their mapping to hardware.Familiarity with AI workloads such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), or transformers.Knowledge of advanced packaging techniques, such as chiplet designs or 3D stacking.Knowledge of advanced memory architectures (e.g., HBM, DDR, GDDR) and their impact on NoC design.What we offer:A competitive salary and benefits package, tailored to recognize your dedication and contributions.The opportunity to collaborate with leading experts in AI and cloud computing, learning from the best and the brightest, fostering continuous growth.An environment that values innovation, collaboration, and mutual respect.Support for personal and professional development, empowering you with the tools and resources to elevate your skills and leave a lasting impact.A pivotal role in the AI revolution, shaping the technologies that power the innovations of tomorrow.About FlexAI:
Founded by
Brijesh Tripathi
and
Dali Kilani , who bring experience from Nvidia, Apple, Tesla, Intel, Lifen, and Zoox, FlexAI is not
just
building a product – we’re
shaping the future of AI.Offices:
Our teams are strategically distributed across three continents—Europe, North America, and Asia—united by a shared mission: to deliver more compute with less complexity.Paris - HQSan Francisco (Bay Area) - US officeBangalore - India officeApply NOW!You’ve seen what this role entails. Now we want to hear from you! Does this opportunity align with your aspirations? If you’re even slightly curious, we encourage you to apply – it could be the start of something extraordinary!At FlexAI, we believe diverse teams are the most innovative teams. We’re committed to creating an inclusive environment where everyone feels valued, and we proudly offer equal opportunities regardless of gender, sexual orientation, origin, disabilities, veteran status, or any other facets of your identity that make you uniquely you.
#J-18808-Ljbffr
FlexAI is at the forefront of revolutionizing AI computing by reengineering infrastructure at the system level. Our groundbreaking architecture, combined with sophisticated software intelligence, abstraction, and an orchestration layer, allows developers to leverage a diverse array of compute, resulting in efficient, more reliable computing at a fraction of the cost.The rapid evolution of machine intelligence has created a need for a new system architecture capable of handling high memory capacity and bandwidth. These are critical bottlenecks in pushing machine intelligence to the next level, where compute demand is expected to increase up to 1000 times current levels.FlexAI has pioneered a groundbreaking solution to tackle these memory challenges. Our innovative compute architecture ensures a well-balanced distribution of memory bandwidth, capacity, and compute density, ensuring maximum utilization of system resources. This architecture is the cornerstone of our datacenter-in-a-box concept, which is enabled by our universal AI compute cloud service. Our hardware solutions are built for seamless deployment with our own AI cloud offerings and other cloud service providers worldwide, setting new standards in performance and efficiency.We are looking for a
NoC Designer
who can architect and design the next generation of Network-on-Chip (NoC), a critical component of our groundbreaking system architecture.Position Overview:
As the
NoC Designer,
you will design and optimize the on-chip communication fabric that connects processing elements, memory hierarchies, and AI accelerators within a System-on-Chip (SoC). Your focus will be on developing scalable and high-performance NoC designs that meet AI-driven applications' bandwidth, latency, and power requirements. You will closely collaborate with SoC architects, AI hardware teams, and software engineers to ensure that the NoC supports the complex data flows required by AI workloads. This is your opportunity to lead, innovate, and usher in a new era of AI hardware design.Success at FlexAI requires an entrepreneurial spirit and startup mindset: the ability to rapidly iterate and make meaningful progress while staying focused on our mission to deliver more compute with less complexity.What you’ll do:NoC Architecture Design:Design and develop high-performance Network-on-Chip (NoC) architectures that interconnect processing cores, AI accelerators, memory subsystems, and peripheral IP blocks.Optimize the NoC for AI-specific workloads such as deep learning inference and training, ensuring that the interconnect can handle high-bandwidth data transfers with low latency.Dataflow and Bandwidth Optimization:Analyze and optimize dataflow patterns for AI workloads, ensuring that the NoC design minimizes congestion, latency, and power consumption while maximizing throughput.Implement advanced techniques such as packet-based routing, dynamic scheduling, and traffic prioritization to manage the data traffic within the SoC efficiently.Scalability and Configurability:Architect scalable NoC designs that support various SoC configurations, from edge AI devices to high-performance AI accelerators.Design modular and configurable NoC components that are easily integrated into different SoC architectures.Performance Modeling and Analysis:Develop performance models and simulation environments to evaluate the NoC architecture under various AI workload scenarios.Perform trade-off analysis between performance, power, and area (PPA) to optimize the NoC design for AI SoCs.Power and Area Efficiency:Work on power-saving techniques such as dynamic voltage and frequency scaling (DVFS), power gating, and clock gating to optimize the NoC for power efficiency.Collaborate with physical design teams to ensure that the NoC design meets area and layout constraints.Cross-Functional Collaboration:Work closely with SoC architects, AI hardware designers, and memory subsystem teams to ensure the NoC design meets system-level requirements.Collaborate with software and firmware teams to optimize communication protocols and ensure efficient data movement across the SoC.What you’ll need to be successful:Bachelor’s, Master’s, or Ph.D. in Electrical Engineering, Computer Engineering, or a related field.5+ years of experience in NoC architecture and design, with a focus on AI, high-performance computing (HPC), or similar workloads.Proven track record in designing and delivering NoC solutions for complex SoCs.Experience with industry-standard NoC IP, network topologies, and interconnect protocols (e.g., AMBA, AXI, CHI).Strong knowledge of NoC architectures, including routing algorithms, flow control, and quality of service (QoS) mechanisms.Experience in performance modeling, simulation, and analysis tools for NoC designs.Proficiency in hardware description languages (e.g., Verilog, VHDL) and NoC design tools.Experience with RTL design, synthesis, and timing closure for custom IP blocks.Strong knowledge of AI hardware architectures, including neural processing units (NPUs), tensor cores, and other AI-specific accelerators.Proficiency in microarchitecture design principles, pipeline optimization, and memory hierarchy design.Experience with performance modeling tools and hardware description languages (e.g., Verilog, VHDL).Familiarity with high-level synthesis (HLS) tools and techniques.Strong analytical and problem-solving skills with a focus on system-level optimization.Ability to work effectively in cross-functional teams, including hardware, software, and verification groups.Excellent communication skills for both technical and non-technical audiences.Preferred Skills:Experience with AI/ML software frameworks (e.g., TensorFlow, PyTorch) and their mapping to hardware.Familiarity with AI workloads such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), or transformers.Knowledge of advanced packaging techniques, such as chiplet designs or 3D stacking.Knowledge of advanced memory architectures (e.g., HBM, DDR, GDDR) and their impact on NoC design.What we offer:A competitive salary and benefits package, tailored to recognize your dedication and contributions.The opportunity to collaborate with leading experts in AI and cloud computing, learning from the best and the brightest, fostering continuous growth.An environment that values innovation, collaboration, and mutual respect.Support for personal and professional development, empowering you with the tools and resources to elevate your skills and leave a lasting impact.A pivotal role in the AI revolution, shaping the technologies that power the innovations of tomorrow.About FlexAI:
Founded by
Brijesh Tripathi
and
Dali Kilani , who bring experience from Nvidia, Apple, Tesla, Intel, Lifen, and Zoox, FlexAI is not
just
building a product – we’re
shaping the future of AI.Offices:
Our teams are strategically distributed across three continents—Europe, North America, and Asia—united by a shared mission: to deliver more compute with less complexity.Paris - HQSan Francisco (Bay Area) - US officeBangalore - India officeApply NOW!You’ve seen what this role entails. Now we want to hear from you! Does this opportunity align with your aspirations? If you’re even slightly curious, we encourage you to apply – it could be the start of something extraordinary!At FlexAI, we believe diverse teams are the most innovative teams. We’re committed to creating an inclusive environment where everyone feels valued, and we proudly offer equal opportunities regardless of gender, sexual orientation, origin, disabilities, veteran status, or any other facets of your identity that make you uniquely you.
#J-18808-Ljbffr