At The San Francisco Tensor Company, we believe the future of AI and high-performance computing depends on rethinking the entire software and infrastructure stack. Today's developers face bottlenecks across hardware, cloud, and code optimization that slow progress before ideas can reach their full potential. Our mission is to remove those barriers and make compute faster, cheaper, and universally portable.
We are building a Kernel Optimizer that automatically transforms code into its most efficient form, combined with Tensor Cloud for adaptive, cross-cloud compute and Emma Lang, a new programming language for high-performance, hardware-aware computation. Together, these technologies reinvent the foundations of AI and HPC.
SF Tensor is proudly backed by Susa Ventures and Y Combinator, as well as a group of angels including Max Mullen and Paul Graham as well as founders and executives of NeuraLink, Notion and AMD. We are partnering with researchers, engineers, and organizations who share our belief that the next breakthroughs in AI require breakthroughs in compute.
We're hiring a Founding GPU Compiler Engineer to build the core compilation infrastructure for our AI compiler. That means taking models from PyTorch, JAX, and TensorFlow and turning them into highly optimized binaries for large-scale AI pre-training.
You'll own the entire compiler stack, from ingesting StableHLO all the way to backend code generation, and you'll work across targets like NVIDIA, AMD, Trainium, and TPU. You'll help shape our architecture, tooling, and overall engineering culture from the very beginning.
Design and implement the main compilation pipeline, from StableHLO to executable GPU and host binaries
Build and extend MLIR dialects and passes to optimize AI workloads
Develop backend code generation for multiple targets (NVIDIA PTX/SASS, AMD GCN/RDNA, Trainium, TPU)
Implement classic compiler optimizations customized for large-scale training (fusion, tiling, memory planning, scheduling)
Build search-based compiler infrastructure to explore different optimization options
Create hybrid codegen paths for cases where direct MLIR lowering isn't practical
Set up testing, benchmarking, and performance regression systems
Work closely with ML researchers to understand workload characteristics and find optimization opportunities
Deep experience with compiler infrastructure (LLVM, MLIR, or similar)
Strong background in GPU architecture and low-level optimization (CUDA, ROCm, or equivalent)
Hands-on experience with at least one of: PTX/SASS, GCN/RDNA assembly, or other GPU ISAs
Familiarity with ML compiler stacks (XLA, TVM, Triton, torch.compile, or similar)
Solid systems programming skills in C++ and/or Rust
Proven track record of building production-grade compiler infrastructure
Background in distributed systems or multi-device compilation
Contributions to open-source compiler projects
Experience with autotuning or search-based optimization
Familiarity with large-scale training infrastructure
Experience with (Stable)HLO
You'll be one of the first engineers defining how we compile and optimize AI workloads. It's a rare chance to build a compiler stack from the ground up, with a direct impact on the efficiency of large-scale AI training.
We believe in the power of in-person collaboration to solve the hardest problems and foster a strong team culture. We offer relocation assistance and look forward to you joining us in our San Francisco office.
The base salary range for this full-time position is $285,000 - $315,000 + bonus + equity + benefits.
AI researchers should be pushing the boundaries of what's possible with new architectures and training methods. Instead, they waste weeks configuring cloud infrastructure, debugging distributed systems, and optimizing their GPU code. We know because we lived it: While training our own models across thousands of GPUs earlier this year, we spent more time fighting our infrastructure than doing actual research.
That's why we're building two things. First, Elastic Cloud: a managed platform that automatically finds the cheapest GPUs across all providers, handles spot instance preemption, and cuts compute costs by up to 80%. Second, automatic kernel optimization that makes training code run faster by modeling hardware topology, often beating hand-tuned implementations.
The problem is that getting high performance across different hardware is genuinely hard. NVIDIA's CUDA moat exists because writing fast kernels requires deep expertise. Most teams either accept vendor lock-in or hire expensive kernel engineers. Our goal is to break the CUDA moat.
The compute bottleneck is the biggest constraint on AI progress. NVIDIA can't manufacture enough GPUs, and their monopoly keeps prices astronomical. Meanwhile, AMD, Google, and Amazon are shipping capable alternative hardware that nobody uses because the software is too hard. We're breaking that moat. If we succeed, anyone will be able to train state-of-the-art models without thinking past their PyTorch code.
Salary
$285,000 - $315,000
Equity
1% - 1.5%
Location
San Francisco
Last stage
Seed
Investors
No applications, no recruiter spam. Just the intro.
A few questions to make sure this role is the right shape for you. Two minutes.
I write the intro, send it to the founder, and handle the back-and-forth.
If they’re a yes, I book the chat. You show up — that’s the whole job-hunt.