MTS & Research: Build Hardware RL for Frontier AI

Collinear AI, Inc.; Sunnyvale (CA); On-site; USD 160,000 - 400,000; Full time

Benefits offered by this job

Equity

Bonus

Competitive salary

Job summary

Collinear is seeking an MTS - Research (Kernels & Hardware RL) to build the data engine for frontier AI. You will develop high-fidelity reinforcement learning environments and evaluation stacks that teach advanced agents to write efficient GPU kernels and solve hardware optimization problems.

You will own environments end to end, from selecting meaningful tasks and reference implementations to building reliable scoring and preventing reward hacking.

Qualifications

• Bachelor’s, Master’s, or PhD in CS, Electrical or Computer Engineering, or demonstrated proof of work through significant open-source contributions or industry experience.
• Strong C/C++ and CUDA or a comparable kernel language, alongside production-quality Python. You have built and profiled kernels and understand memory hierarchy, data movement, parallelism, occupancy, and numerical correctness.
• ML Fluency with hands-on experience with LLMs and evaluation/optimization understanding. Translate systems problems into executable tasks with clear feedback.
• Empirical mindset: experience conducting reproducible performance experiments, inspecting profiler results, and distinguishing real improvements from measurement noise or benchmark shortcuts.

Responsibilities

• Build Agentic Environments: design diverse tasks for GPU and accelerator kernels, low-level systems programming, and hardware-aware optimization.
• Programmatic and Agentic Verification: create reproducible correctness tests and performance graders, benchmark against trusted implementations, secure sandboxes against timing exploits and rewards hacking.
• Close the Loop: analyze agent trajectories, curate high-signal tasks and data, and conduct post-training experiments to improve real capability across workloads and hardware.
• Collaborate: work daily with founders and research staff to shape the roadmap and push state-of-the-art in AI reliability.
• Scale the Infrastructure: build reliable compilation, execution, profiling, and evaluation pipelines supporting repeated experiments across targets.

Skills

C/C++

CUDA

Python

GPU kernels

Performance profiling

Education

Bachelor’s, Master’s, or PhD in CS/ Electrical/ Computer Engineering

Tools

Triton

CUTLASS

LLVM/MLIR

RTL/HDL

Job description

Collinear is seeking an MTS - Research (Kernels & Hardware RL) to build the data engine for frontier AI. You will develop high-fidelity reinforcement learning environments and evaluation stacks that teach advanced agents to write efficient GPU kernels and solve hardware optimization problems.

You will own environments end to end, from selecting meaningful tasks and reference implementations to building reliable scoring and preventing reward hacking.