Middle/Senior Python Developer

We're hiring RL Environments Engineers to design and build MLE/SWE environments that
deliver high-quality, diverse tasks with minimal supervision. You will target a specific language
model, meet a defined difficulty distribution, and deliver about one task every 10 hours. 

The goal is to develop RL environments that teach LLMs low-level programming, kernel development, and CPU/GPU optimization.
The job requires independence and engineering responsibility.

Requirements:
- Strong Python (engineering code, not just notebooks);
- Real-world experience with LLM/GenAl in production;
- Experience with end-to-end pipelines;
- Docker and production thinking;
- Ability to handle expected throughput and respond quickly;
- Candidates with experience in one or more of the following areas are welcome;
- Memory architecture and execution models;
- Multithreading and concurrency;
- JIT/AOT compilation (Triton, LLVM, MLIR, TVM, etc.);
- Kernel frameworks (CUDA, CUTLASS, HIP, ROCm);
- Modern C++ and low-level optimization;
- PyTorch custom operators and backend integrations;
- Mixed/low-precision computing;
- English level C1/C2.

Responsibilities:
- Design and development of environments and tasks;
- Working with a given model and complexity distribution;
- Approximately 1 task per ~10 hours;
- Quickly make edits based on client feedback.

It will be a plus:
- Experience with RL environments or evaluation frameworks;
- Work in regulated or high-stakes domains;
- Experience with MLOps, CI/CD, and monitoring;
- Systems background (C++ / Rust / Java / Scala);
- Understanding of RL, bandits, and agent systems.

Working conditions:
- Remote, contract model;
- Project duration: 1 year or more;
- At least 4 hours of PST (UTC-8, the winter time zone used on the West Coast of the US, Canada, and Mexico);
- Quick onboarding, start from day one.