Firmware Automation Engineer
Job description
A remote EU-based role focused entirely on silicon and low-level software layers, designing and maintaining automation frameworks for firmware updates across a global fleet of servers, switches, and GPU clusters. The role builds safety nets (health checks, canary deploys, automated rollbacks) to update hardware rapidly without taking down the cloud.
At a glance
- Employment
- Full-time · Employee (employment agreement)
- Work mode
- Remote · EU
What you'll do
- Design and execute automated, zero-touch firmware update pipelines across a global fleet of servers, switches, and accelerators
- Own the firmware lifecycle for multiple hardware profiles, including Systemboards (BIOS/UEFI), BMCs, InfiniBand HCAs, Network Switches, and high-performance GPUs
- Leverage Redfish and IPMI APIs to programmatically discover, configure, inventory, and update bare-metal assets at scale
- Architect robust telemetry pipelines to assess hardware health post-flash, and build automated rollback mechanisms to recover from faulty vendor payloads
- Integrate firmware management pipelines into existing configuration management ecosystems (SaltStack/Ansible) and provisioning workflows
- Deep-dive into hardware errata and bug reports, collaborating with hardware vendors and internal platform teams to patch critical CVEs and performance regressions
What we're looking for
- Proven experience managing and updating firmware across thousands of nodes simultaneously; writing code to update clusters rather than one-off flashes
- Mastery of the DMTF Redfish specification for server management, out-of-band communication, and telemetry collection
- Strong understanding of architectural differences in managing firmware across systemboards, network stacks (InfiniBand/Ethernet), and GPU architectures
- A paranoid mindset regarding hardware health: verifying if a device is truly healthy post-boot and safely executing an unattended rollback if it isn't
- Proficiency in Python, Go, or advanced shell scripting alongside configuration management tools (Ansible/Salt) to interact with hardware APIs
Nice to have
- Experience working within the OpenBMC ecosystem or building custom BMC firmware images
- Deep operational familiarity with NVIDIA/Mellanox firmware management tools (mstflint, mlxup)
- Experience managing firmware lifecycles within large-scale AI/ML training clusters or high-performance computing (HPC) environments
Benefits
An opportunity to make a clear impact and work alongside world-class engineers, researchers, and partners across the global AI ecosystem
- Cash + equity compensation along with various fringe benefits
- Profitable operations with rapid, sustained growth
- 31 nationalities, with 6 different ones on the management team
Apply
About Verda
A fully featured European AI cloud
Verda is building a fully featured European AI cloud with everything needed to train, experiment with, and deploy AI models. Its GPUs run on 100% renewable energy. The company describes itself as ambitious, curious, and gutsy doers with low hierarchy and high morale in teams.
Mission
Building a fully featured European AI cloud with instant, low-cost access to intelligence, with GPUs running on 100% renewable energy.
Key facts
- GPUs run on 100% renewable energy
- Profitable operations with rapid, sustained growth
- 31 nationalities, with 6 different ones on the management team
What they value
- Ambitious
- Curious
- Gutsy doers
- Low hierarchy
- High morale