Runtime Engineer
A San Francisco venture capital firm and builder community investing in technically rigorous frontier-technology companies.
About Pebblebed
Pebblebed is a venture capital firm, builder community, AI research lab, and event space focused on technically rigorous frontier technology. Its current portfolio emphasizes AI, robotics, developer infrastructure, biotech, and related frontier technologies.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
Design and build a multi-target runtime for an AI compiler stack, making compiled workloads execute efficiently, correctly, and at scale across diverse hardware targets. The role involves low-level parallelization, kernel scheduling, performance analysis, prototyping runtime ideas, benchmarking compiler outputs, and collaborating with compiler and product teams.
Requirements
- BS degree in Computer Science, Computer Engineering, or equivalent practical experience.
- 4+ years of experience working with compilers or runtime systems.
- Deep understanding of asynchronous and concurrent programming.
- 4+ years of experience with C/C++ (C++14 or newer).
- Understanding of hardware architecture, including vector and scalar registers, instructions, and memory hierarchies.
- Knowledge of operating system kernel development or hypervisor development.
- Preferred: Master's or PhD in Computer Science, Computer Engineering, or equivalent.
- Preferred: Experience with CUDA or ROCm GPU compute libraries.
- Preferred: GPU programming and optimization experience.
- Preferred: Background in high-performance computing.
- Preferred: Knowledge of PyTorch, JAX, or Triton.
- Preferred: Experience programming large compute clusters.
Responsibilities
- Design, develop, maintain, and improve the multi-target runtime.
- Apply parallelization and partitioning techniques to automate kernel generation and exploit optimized execution paths.
- Rapidly prototype and conduct data-driven exploration of new runtime ideas.
- Benchmark and analyze optimizing compiler outputs on target hardware.
- Build tools to collect and analyze performance bottlenecks.
- Collaborate with product teams to understand ML engineer needs and improve runtime architecture.
Benefits
- Equity
- Company bonus opportunities
- Medical coverage
- Dental coverage
- Vision coverage
- Retirement savings plan
- Supplemental wellness benefits