Senior Principal Engineer, NPU Compiler

CARIAD, Inc.•Mountain View, CA
•$247,000 - $312,000•Hybrid

About The Position

The Senior Principal Engineer, NPU Compiler provides technical leadership for CARIAD’s NPU inference compiler within the NPU Hardware and Software organization. The role leads compiler architecture, implementation standards, and release-readiness criteria within the approved NPU architecture. It translates supported machine-learning workloads into efficient target code and works with model enablement, kernel, runtime, firmware, verification, and hardware teams to define interfaces, validate numerical behavior, and improve performance. The engineer guides complex compiler initiatives, mentors engineers, and delivers maintainable compiler releases with documented functionality, limitations, and performance. CARIAD is an automotive software development team within the Volkswagen Group. Its mission is to build a unified software platform and digital products for Volkswagen Group brands, with a focus on making the automotive experience safer, more sustainable, more comfortable, more digital, and more fun. The position is based in Mountain View, California, with in-office presence three to five days per week and some hybrid flexibility. No travel is planned at this time. The salary range in Mountain View is $247,000–$312,000, dependent on factors such as geographical differentials, credentials or certifications, industry-based experience, qualifications, and training. Employment is subject to export control and sanctions compliance screening.

Requirements

  • 13 or more years of experience in compiler development or architecture, or equivalent relevant experience, particularly targeting AI or machine-learning hardware accelerators.
  • Bachelor’s degree in Computer Science, Computer Engineering, or Electrical Engineering, or equivalent practical experience.
  • Strong understanding of machine-learning algorithms and their computational implications.
  • Hands-on experience with at least one compiler infrastructure such as TVM, IREE, XLA, MLIR, or LLVM.
  • Strong proficiency in C++ and Python for compiler development.
  • Experience with compiler graph optimization and memory-management strategies.
  • Ability to translate high-level functional requirements into detailed technical designs.
  • Deep knowledge of hardware architecture principles and AI-accelerator design concepts.
  • Demonstrated delivery of compiler architecture from technical design through validated releases, with influence across hardware and software teams.
  • Experience with neural-network quantization, precision selection, numerical-error analysis, and accuracy regression in collaboration with model-deployment and hardware teams.
  • Experience with compiler testing and reproducible releases, including reference comparisons, memory-safety checks, and performance-regression diagnosis.
  • Ability to communicate complex technical concepts clearly and provide technical leadership and mentorship across engineering teams.

Nice To Haves

  • Prior experience with NPU hardware.
  • Knowledge of automotive standards and Functional Safety requirements.
  • Background in high-performance computing and parallel-processing architectures.
  • Publications or contributions to open-source compiler projects.
  • Familiarity with generative-AI-assisted engineering tools and the ability to evaluate their benefits, limitations, and appropriate use.
  • Commitment to continuous learning as compiler technologies and development practices evolve.
  • Master’s degree in Computer Science, Computer Engineering, or Electrical Engineering.

Responsibilities

  • Design and evolve NPU compiler architecture to support approved hardware capabilities, compiler/runtime interfaces, and target workloads.
  • Implement and evaluate hardware-independent and hardware-specific graph-optimization passes.
  • Develop and optimize tiling and memory-management algorithms to use NPU resources efficiently.
  • Create advanced optimization passes for approved automotive neural-network inference workloads.
  • Define model import and operator coverage, quantization semantics, fallback policy, and graph-to-target lowering.
  • Map high-level operations to optimized library macros and convert them into hardware-level instructions.
  • Generate and manage DMA commands for data movement and execution within the hardware platform.
  • Collaborate with hardware engineers and system architects to validate NPU integration and improve performance against agreed workload metrics.
  • Implement compile-time scheduling and resource allocation, and define runtime queueing, admission, synchronization, and recovery interfaces with the NPU software lead.
  • Specify target code generation, binary formats, compiler/runtime ABI, memory layout, DMA and synchronization contracts, and compatibility across hardware revisions.
  • Define and maintain compiler-readiness criteria and reference-output comparisons for numerical correctness, operator and model coverage, and performance; document unsupported cases, limitations, and fallback costs.
  • Evaluate compilation techniques against committed inference workloads and deliver validated improvements into maintained compiler releases.
  • Lead complex compiler initiatives, set technical direction within the approved NPU architecture, and align engineering teams and stakeholders on priorities, tradeoffs, and delivery criteria.
  • Establish compiler-engineering standards for design review, testing, reproducible builds, and release quality.
  • Mentor engineers on compiler architecture, code generation, numerical correctness, and effective engineering practices.
  • Guide compiler implementers and kernel engineers through design reviews, reproducible builds, and release planning; coordinate runtime and model-enablement work with their respective owners.
  • Use internal and external trends in compiler technology and machine learning to guide decisions and recommend how advancements can be applied at CARIAD.
  • Communicate complex technical concepts across global, cross-functional teams; build collaboration and alignment without relying on formal people-management authority.
  • Manage complex cross-functional and cross-team initiatives, balance competing technical priorities, clarify decision ownership, and maintain productive relationships with partner engineering organizations.
  • Collaborate effectively across geographies and time zones and apply advanced analytical and systems thinking to compiler and hardware/software integration challenges.

Benefits

  • Performance-based merits and annual bonus.
  • Medical, dental, and vision benefits.
  • 401(k) with employer match and defined contribution plan.
  • Short- and long-term disability insurance.
  • Basic life and AD&D insurance.
  • Employee assistance program.
  • Tuition reimbursement and student loan repayment plans.
  • Maternity and non-primary caregiver leave.
  • Adoption assistance.
  • Employee referral program.
  • Vacation and paid holidays.
  • Vehicle lease program covering registration and insurance fees.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service