MLOps Engineer

Your tasks

• Design challenging technical tasks involving debugging, profiling, kernel optimization, and model inference.
• Evaluate AI-generated solutions for technical accuracy and performance.
• Develop guidelines and scoring rubrics for ML systems tasks.
• Provide detailed technical feedback to research and engineering teams.
• Work with experts to improve the quality and consistency of AI training data.

Requirements

• At least 2 years of professional experience in ML systems, infrastructure, model serving, or accelerator performance engineering.
• Hands-on experience with JAX or PyTorch.
• Knowledge of tools such as CUDA, Triton, Pallas, or CUTLASS.
• Experience with performance profiling, debugging, or distributed workloads.
• Understanding of large-scale model serving tools such as vLLM, SGLang, or TensorRT-LLM.
• Experience with modern accelerators such as TPUs, NVIDIA GPUs, Trainium, or MI300.
• Strong analytical and written communication skills.
• Ability to commit to at least 40 hours per week on weekdays.