BMC Developer

About Tensor Wave

Our mission is simple: deliver seamless, secure, reliable, and resilient AI compute 've built a versatile cloud platform that eliminates infrastructure barriers, empowering builders to focus on innovation instead of fighting their stack. Because breakthrough AI should move at the speed of ideas, not infrastructure.

We are seeking a highly skilled Staff BMC Developer to own the software lifecycle, configuration, integration, and long-term management of Tensor Wave’s custom Baseboard Management Controller systems.

This role will serve as the technical owner for our out-of-band management architecture across next‑generation AI compute platforms. The primary focus will be writing software for, configuring, integrating, and maintaining Axiado 3000‑series BMC modules, with specific emphasis on OpenBMC, Redfish, PLDM, MCTP, low‑level hardware interfaces, and integration with AMD Universal Base Board architecture for high‑density GPU platforms.

The Staff BMC Developer will work closely with infrastructure engineering, hardware provisioning, network engineering, platform engineering, AMD engineering, and OEM hardware vendors to ensure our server management layer is stable, secure, observable, automatable, and production‑ready at fleet scale.

This role is not a general firmware support position. It is a senior technical ownership role responsible for the management‑plane software foundation required to deploy, monitor, update, recover, and operate large‑scale AI compute infrastructure.

• Own the full software lifecycle for Tensor Wave's custom BMC systems: firmware development, image customization, board configuration, validation, release management, upgrade workflows, and recovery procedures.

• Lead development and deployment for Axiado 3000-series BMC modules (Smart‑SCM
  3002 / DC‑SCM architectures).

• Define BMC platform standards covering firmware builds, configuration management, access control, secure interfaces, and automation integration.

• Build, customize, maintain, and troubleshoot OpenBMC‑based firmware for Tensor Wave‑specific hardware platforms.

• Integrate platform‑specific sensors, inventory, and control paths; support board bring‑up and hardware enablement.

• Develop in C, C++, Python, and Shell within Yocto/Bit Bake and OpenBMC build systems.

• Lead integration and operationalization of Axiado BMC modules; configure, debug, and validate sensor, power, reset, and firmware update workflows.

• Architect BMC integration with AMD Universal Base Board platforms supporting high‑density MI‑series GPU systems.

• Coordinate across BMC, BIOS, GPU, NIC, and system firmware dependencies; work directly with AMD, Axiado, and OEM vendors.

• Debug and configure hardware communication interfaces: I2C, I3C, SPI, UART, PCIe, IPMI, PLDM, MCTP, and Redfish/DMTF APIs.

• Design and maintain robust firmware update processes (BIOS, BMC, NIC, GPU, CPLD/FPGA), including PLDM‑based workflows, rollback, version tracking, and fleet‑scale rollout planning.

• Implement and maintain Redfish APIs exposing power, thermal, inventory, sensor, and health data for automation and bare‑metal provisioning.

• Ensure platform health metrics (thermal, power, voltage, fan, GPU, system health) are exposed to infrastructure and monitoring systems — not trapped in firmware or vendor tools.

• Serve as primary technical point of contact with AMD, Axiado, OEM vendors, and ODM platform teams; lead escalations and validate vendor fixes before production.

• Partner closely with Infrastructure, Platform Engineering, Dev Ops, Network Engineering, Observability, and Security teams.

• Deep hands‑on experience with BMC firmware development and OpenBMC (Yocto/Bit Bake).

• Strong proficiency in C, C++, Python, and Shell scripting in embedded Linux environments.

• Working knowledge of hardware protocols: I2C, I3C, SPI, UART, PCIe, IPMI, PLDM, MCTP.

• Experience with Redfish/DMTF APIs and out‑of‑band server management.

• Ability to work close to the metal…