The reward signal here is unusually clean for RL: measure wall-clock on real silicon gated by a formal proof. No proxy metrics, no reward hacking that survives contact with the verifier. The system already does things worth seeing. Given only a naive attention spec, the search found FlashAttention v4-level performance on a B300 in about 25 minutes, but it also generalizes. The same loop runs on AMD, TPU and Trainium as well as targets we'd never seen before, such as Apple Silicon.

You'll have access to better tooling than anywhere else because we own the stack all the way down to the ISA, allowing us to create kernels that others can't even express. For example, our custom LLVM backend emits cubins directly, without ptxas, letting us work on instruction selection, scheduling, register allocation and stall counts. Our team understands the hardware better than anyone else, to this extent we've built a bit-exact software model of Blackwell's tcgen05.

The kernels you create will ship immediately on runs such as pre-training AlphaFold v3 at 3.4× the throughput, post-training robotics models on Trainium or running our custom RL rollout engine on TPU at multi-100B parameter scale.

What You'll Do

• You'll design and implement RL systems that search over a massive program space: instruction selection, schedules, tile sizes, fusion strategies and phase ordering

• You'll build agentic compilation loops that use LLMs to reason about IR, propose transformations and learn from measured results

• You'll design the search and credit-assignment machinery around an exact reward built from measured latency on real hardware and a formal correctness proof

• You'll create representations and embeddings of compiler IR that hold up under learned optimization

• You'll build the training infrastructure for compiler optimization agents, including rollout throughput and distributed evaluation on real silicon

• You'll push transfer and cold-start performance so that the search works on unfamiliar targets and new ISAs from the first trial

• You'll close the loop with production workloads so that the compiler keeps improving from what customers actually run

• You'll work directly with our compiler and kernel engineers to land learned components in the shipping pipeline

• You'll run rigorous experiments, dig into the results, iterate on them and publish or open-source some of them

What We're Looking For

• Someone with a strong background in reinforcement learning with hands-on experience training agents

• Someone with experience building LLM agents, tool use and other agentic systems

• Someone with familiarity in GPU programming concepts and a willingness to get down to the ISA

• Someone proficient in PyTorch or JAX

• Someone with the ability to design and run experiments that produce trustworthy results

Nice to Have

• Someone with experience in ML compiler stacks (XLA, TVM, Triton, MLIR) or LLVM backends

• Someone with a background in program synthesis, superoptimization, combinatorial search, formal methods, SMT solvers or verified compilation

• Someone familiar with RLHF, reward modeling or preference learning

• Someone with research contributions in RL or learned optimization

• Someone familiar with GPU performance optimization, profiling and microbenchmarking

Why Join Us

Compiler optimization is the rare place where a learned system gets an enormous action space and a ground-truth reward at the same time. What usually blocks this work is everything around it (a background that can't execute what the search finds or no way to trust the result). You get both here: direct cubin emission below ptxas and a prover that tells you whether the program is correct. In this role, you'll have the tools and researches to let your most ambitious research ideas run wild and if they find valid kernels, ship them into production in days rather than becoming a paper and then nothing.

We're a small team operating at frontier scale. We pre-trained foundation models on 4,000 AMD GPUs as a team of three, designed and brought up GB300 NVL72 clusters and designed a TOP500 supercomputer.

We believe that hard problems get solved in person and most of our work happens at our office in San Francisco. We offer relocation assistance and, where possible, we'd like you here as often as possible.

The base salary range for this full-time position is $275,000-$315,000, plus meaningful equity and benefits.

A meeting lands on your calendar

If they’re a yes, I book the chat. You show up — that’s the whole job-hunt.