In this episode of the Embedded Frontier podcast, host Jacob interviews Darwin from GitLab's field CTO office about the adoption and implementation of DevOps practices in embedded systems development. They explore the unique challenges embedded developers face when modernizing their workflows, including managing complex codebases with hundreds of millions of lines of code, compliance requirements, and the critical differences between software-only products and embedded systems where software is just one component of the final product.
Key Takeaways:
• Embedded systems require different DevOps approaches than pure software products since shipping software doesn't mean shipping the final product
• Modern vehicles contain 650 million lines of code in 2025, up from 200 million just five years ago, creating new complexity management challenges
• Three categories of embedded systems each need different DevOps strategies: digital disruptors, stable machines, and functional safety systems
• Containerized builds and shared development environments eliminate "works on my machine" problems and create reproducible, auditable builds
• Software supply chain security through Solza attestation provides traceability from source code to final artifacts
• Compliance as code can automate many regulatory requirements like ISO 26262 and MISRA C++, reducing manual bottlenecks
• AI integration at the platform level helps embedded developers onboard to DevOps without becoming DevOps experts
• Continuous delivery (creating release-ready firmware) is more appropriate for embedded than continuous deployment to production
• Automated testing and QA are crucial to prevent manual processes from becoming the limiting factor in development speed
• Over-the-air updates in embedded systems require managed deployments with higher reliability than cloud container replacements
