Systems Software Engineer, NPU Management Interface
FuriosaAI · Seoul, South Korea ·
- Category
- Software engineering
- Company size
- 11-50
Develops firmware/software for NPU management interfaces, handling communication between host systems and AI accelerators via protocols like MCTP/PLDM, while ensuring telemetry, health monitoring, and compliance with industry specs.
About FuriosaAI
FuriosaAI builds high-performance, high-efficiency AI compute for the Inference Era. Founded in 2017 by veteran semiconductor and AI algorithm engineers, Furiosa operates globally with offices in Korea and Silicon Valley, along with a compiler-focused R&D lab in Lisbon.
Our vision is to make AI computing sustainable, enabling access to powerful AI for everyone on Earth. We solve the AI hardware energy and operational cost crisis at the architectural level, rather than through brute force, building the world's first truly AI-native compute platform to unlock the full potential of artificial intelligence for every enterprise.
Key Responsibilities
-
Design and develop the NPU Management Interface (MI) firmware/software enabling communication between Host/BMC and NPU devices
-
Implement and maintain MCTP, PLDM, and custom MI command handling for out-of-band NPU management, monitoring, and control
-
Develop device-management features over SMBus/I²C, I3C, PCIe VDM, or custom sideband channels
-
Integrate MI functionality into the NPU firmware, including:
-
Health and error reporting
-
Thermal and power telemetry
-
Runtime status, utilization metrics, and debug information
-
-
Ensure compliance with industry specifications by performing spec-driven design, implementation, and validation
-
Support bring-up, interoperability testing, rack-scale platform integration, and system-level validation
-
Develop test strategies and validation tools based on MCTP and PLDM specifications
-
Perform protocol compliance testing, regression testing, and interoperability verification
Minimum Qualifications
-
Strong proficiency in embedded C or C++
-
Experience with firmware development for NPU/accelerator, GPU, or SoC
-
Understanding of management protocols including MCTP (over I²C/SMBus, I3C or PCIe VDM) and PLDM
-
Experience with low-level interfaces: SMBus/I²C, I3C, SPI, PCIe
-
Ability to interpret complex protocol specifications and convert them into robust implementations
-
Familiarity with device telemetry, sensor frameworks, watchdog/reset flows, and health monitoring
-
Experience with system-level debugging using logic/protocol analyzers and low-level debug tools
-
Knowledge of embedded systems, bare-metal or RTOS environments, and firmware lifecycle flows
Preferred Qualifications
-
Experience of BMC firmware stacks such as OpenBMC, Redfish, IPMI, and PLDM device-model implementations
-
Background in spec creation, requirement definition, or standards compliance validation
-
Experience defining FRU data, power/thermal management policies, and diagnostics frameworks
-
Familiarity with secure provisioning, firmware update mechanisms, and lifecycle state management
-
Experience with large-scale datacenter or HPC system integration (rack-level management, telemetry aggregation)
-
Contributions to firmware for accelerator, MCTP/PLDM implementations, or open-source system firmware projects
Why Join FuriosaAI
The defining bottleneck of the AI era is building the right hardware and software stack to run it at global scale. Furiosa is solving this challenge holistically from the ground up.
With our flagship chip, RNGD, in mass production today and our next-generation platform in development with Broadcom, we are proving that full-stack, tensor-native compute is the future of AI infrastructure. This is a pivotal moment to join our team, right as we accelerate our global expansion.
At Furiosa, you will:
Solve AI’s Most Urgent Challenge. Help build the high-performance, energy-efficient inference hardware and software required to fulfill the promise of advanced AI.
Pioneer Full-Stack Co-Design. Work with teams that are architecting solutions from silicon up through the compiler (featuring innovations like Tensor Contraction Language and Virtual ISA) and serving frameworks.
Ship Real-World Silicon, Software, and Solutions. Turn breakthrough technology into commercial deployment. RNGD is in mass production with TSMC and running live enterprise workloads for global leaders like LG AI Research and Samsung SDS.
Partner With the Industry's Best. Collaborate across an elite global ecosystem that includes TSMC, Broadcom, SK Hynix, and GUC.
Do Your Life’s Best Work. Join a brilliant, low-ego, mission-driven team in a high-trust environment that values autonomy, intellectual curiosity, and shared ambition.
Contact