KA
System Firmware Architect(AI Memory Fabrics) – CXL/PCIe Memory fabric controller
Accepting applicationsKandou AI · Pune District, Maharashtra, India
Full-Time Principal AIC++DDRPCIePCie
Estimated market salary
₹34-61 LPA
This is a SiliconBoard market estimate, not an employer-posted salary.
Posted
21h ago
Category
Design
Experience
Principal
Country
India
System Firmware Architect (AI Memory Fabrics) – CXL/PCIe Memory fabric controller
Location: Greater Bengaluru Area (Hybrid)/ Greater Hyderabad /Pune
https://www.thecube.net/events/nyse/ai-factory-data-center-of-the-future/content/Videos/d1ad8af3-1424-4336-a9b3-1d02b5a354b4
we are redefining the economics of AI infrastructure. Our mission is to democratise AI by significantly reducing the Total Cost of Ownership (TCO) of hardware systems — a critical barrier to scalable adoption.
Our proprietary MIMO-over-copper technology powers a high-performance, chiplet-based AI memory fabric that is both scalable and energy-efficient. Unlike traditional interconnects, our solution reduces power consumption significantly while preserving high bandwidth and ultra-low latency — unlocking unprecedented efficiency for AI training and inference at scale.
our architecture is not just an incremental improvement — it’s a foundational shift in how AI hardware is built for the future.
. Location: Hyderabad/Bangalore
System Firmware Architect (AI Memory Fabrics) – CXL/PCIe Memory fabric controller
The responsibility of this role is to architect and develop firmware for Kandou’s Memory fabrics products
Firmware is responsible for implementing and supporting the major differentiating features of Kandou’ products.
Basic qualification
Strong academic and technical background in Electronics & Computer engineering. At a minimum, a Bachelor’s in EE or Computer Science is required, and a Master’s is preferred.
Experience developing firmware (Bare metal/RTOS) to execute in on-chip microcontrollers/processors
Minimum 15+ years’ (Lead/Architect) experience supporting/developing complex SoC/Silicon products for Server, Storage, and/or Networking applications
Required experience
Hands-on experience designing and developing embedded firmware for PCIe/CXL/NVME protocols with proficiency in C (preferred) or C++, including development of C-based SDKs.
Experience with RTOS/Zephyr OS firmware development and enablement
Deep understanding of SoC interfaces to common IP blocks such as PCIe Controllers, NVME Controllers, DDR controllers, AMBA/AHB interfaces, on-chip memory interfaces, and other similar interfaces
Work with various cross-functional groups such as architect, other development teams, QA/validation, product management and customer teams across various sites
Experience in mentoring and train the team
Preferred experience
Experience with storage or SSD firmware, including NVMe, PCIe, CXL or flash management concepts.
Experience with industry forums and collaboration workgroups such as OCP and OpenBMC
Hands-on experience with simulation or virtual platform–based development (e.g., Renode, QEMU, or similar)
Skills
CXL
Memory controller
FAbric
"bare metal"
PCie
NVMe
C
RTOS
SoC
"DDR controllers"
SSD
OCP
OpenBMC
Renode
QEMU
Zephyr
Contact
Uday
muday_bhaskar@yahoo.com
Mulya Technologies
"Mining the Knowledge Community"
Show more Show less
Location: Greater Bengaluru Area (Hybrid)/ Greater Hyderabad /Pune
https://www.thecube.net/events/nyse/ai-factory-data-center-of-the-future/content/Videos/d1ad8af3-1424-4336-a9b3-1d02b5a354b4
we are redefining the economics of AI infrastructure. Our mission is to democratise AI by significantly reducing the Total Cost of Ownership (TCO) of hardware systems — a critical barrier to scalable adoption.
Our proprietary MIMO-over-copper technology powers a high-performance, chiplet-based AI memory fabric that is both scalable and energy-efficient. Unlike traditional interconnects, our solution reduces power consumption significantly while preserving high bandwidth and ultra-low latency — unlocking unprecedented efficiency for AI training and inference at scale.
our architecture is not just an incremental improvement — it’s a foundational shift in how AI hardware is built for the future.
. Location: Hyderabad/Bangalore
System Firmware Architect (AI Memory Fabrics) – CXL/PCIe Memory fabric controller
The responsibility of this role is to architect and develop firmware for Kandou’s Memory fabrics products
Firmware is responsible for implementing and supporting the major differentiating features of Kandou’ products.
Basic qualification
Strong academic and technical background in Electronics & Computer engineering. At a minimum, a Bachelor’s in EE or Computer Science is required, and a Master’s is preferred.
Experience developing firmware (Bare metal/RTOS) to execute in on-chip microcontrollers/processors
Minimum 15+ years’ (Lead/Architect) experience supporting/developing complex SoC/Silicon products for Server, Storage, and/or Networking applications
Required experience
Hands-on experience designing and developing embedded firmware for PCIe/CXL/NVME protocols with proficiency in C (preferred) or C++, including development of C-based SDKs.
Experience with RTOS/Zephyr OS firmware development and enablement
Deep understanding of SoC interfaces to common IP blocks such as PCIe Controllers, NVME Controllers, DDR controllers, AMBA/AHB interfaces, on-chip memory interfaces, and other similar interfaces
Work with various cross-functional groups such as architect, other development teams, QA/validation, product management and customer teams across various sites
Experience in mentoring and train the team
Preferred experience
Experience with storage or SSD firmware, including NVMe, PCIe, CXL or flash management concepts.
Experience with industry forums and collaboration workgroups such as OCP and OpenBMC
Hands-on experience with simulation or virtual platform–based development (e.g., Renode, QEMU, or similar)
Skills
CXL
Memory controller
FAbric
"bare metal"
PCie
NVMe
C
RTOS
SoC
"DDR controllers"
SSD
OCP
OpenBMC
Renode
QEMU
Zephyr
Contact
Uday
muday_bhaskar@yahoo.com
Mulya Technologies
"Mining the Knowledge Community"
Show more Show less