Develop and shape a platform for deploying AI applications at scale, owning core distributed-systems components and improving infrastructure performance, scalability, and reliability.
Software Architect
📜 Description
- Architect the CMIS 5.x stack, including DataPath State Machine, VDM, I2C/MCTP, and firmware-software interfaces.
- Lead hyperscaler host interoperability testing and build test matrices and regression suites.
- Develop manufacturing test orchestration, including ATE integration, yield analytics, and on-chip stimulus.
- Build firmware-to-host and cloud telemetry pipelines, including formats, compression, and SDK APIs.
- Own customer SDKs and tools, release management, and log analysis.
- Ensure compliance with CMIS 5.x, OSFP MSA, SFF-8024, and OIF standards.
🛠️ Requirements
- Good knowledge of C/C++, Python, systems language (Go/Rust); CI/CD, conformance tests
- Good knowledge of I2C/MCTP, SFF-8024, vendor extensions
- Good understanding of standards implementation (OIF, SFF, IEEE mgmt-plane)
- Good knowledge of architecture documentation (SHALL/MUST, REQ traceability)
- Self-motivated, with strong sense of ownership and responsibility
- Good communication and reporting skills
- Principal / Senior Staff (8-10 years post-degree)
- 8-10 years high-speed interconnect software (module SW, management, test, SDK)
- CMIS 4.x/5.x implementation and OSFP MSA experience
- Hyperscaler interop with NIC/switch ASICs (Nvidia, Broadcom)
Full job description
At Kandou, 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.
Kandou’s architecture is not just an incremental improvement — it’s a foundational shift in how AI hardware is built for the future.
We are actively seeking a Software Architect based in one of the following locations: India (Hyderabad/Bangalore), UK (Reading/Cambridge) Switzerland (Lausanne).
Responsibilities:
Architect CMIS 5.x stack (DataPath State Machine, VDM, I2C/MCTP, FW-SW interface)
Lead host interop tests for hyperscalers (Meta, AWS, etc.), build test matrix and regression suite
Develop manufacturing test orchestration (ATE integration, yield analytics, on-chip stimulus)
Build telemetry pipeline from FW to host/cloud (formats, compression, SDK APIs)
Own customer SDK and tools (C/Python, CLI, log analysis), release management
Ensure standards compliance (CMIS 5.x, OSFP MSA, SFF-8024, OIF)
Travel: Up to 25% (customer engagement, manufacturing, standards)
Skills:
Good knowledge of C/C++, Python, systems language (Go/Rust); CI/CD, conformance tests
Good knowledge of I2C/MCTP, SFF-8024, vendor extensions
Good understanding of standards implementation (OIF, SFF, IEEE mgmt-plane)
Good knowledge of architecture documentation (SHALL/MUST, REQ traceability)
Self-motivated, with strong sense of ownership and responsibility
Good communication and reporting skills
Experience:
Principal / Senior Staff (8-10 years post-degree)
8-10 years high-speed interconnect software (module SW, management, test, SDK)
CMIS 4.x/5.x implementation and OSFP MSA experience
Hyperscaler interop with NIC/switch ASICs (Nvidia, Broadcom)
Test software architecture with ATE/yields
Manufacturing and field support experience
Education:
BS in Electrical Engineering, Computer Engineering, or Computer Science (MS preferred)
If this is the role you have been looking for and you want to be part of a growing Company, with an exciting future then we would really love to hear from you. Together We Kandou It!
Similar jobs
Search more Software Engineer jobsDevelop embedded robot software that improves real-time field operations, fleet reliability, and field-to-cloud data flow, while integrating hardware and resolving issues across deployed devices.
Design and optimize C++ software for autonomous vehicle ride states and rider-facing features, integrating vehicle systems and cloud services while improving performance, reliability, and safety.
Leads integration, verification, and reliability work for automated NGS diagnostic systems, combining hands-on assay testing with validation strategy, risk analysis, design-control documentation, and technical mentorship.
Manage and improve analog and digital EDA environments, developing automated chip-design flows, supporting engineers and vendors, and enabling reliable use of tools and PDKs across advanced technology nodes.
