Senior Research HPC Engineer (Language AI)

TL;DR

Senior Research HPC Engineer (Language AI): Building and optimizing high-performance computing solutions for Language AI technology with an accent on efficient parallelization on GPU clusters and multimodal LLM inference. Focus on profiling, debugging, and optimizing GPU kernels and low-level systems configuration for state-of-the-art models.

Location: This role can be based out of the following locations: Munich, Cologne, London. We offer a hybrid work schedule, with team members coming into the office twice a week.

Company

hirify.global is a global communications platform powered by Language AI, building products to transform communications and break down language barriers for businesses and individuals worldwide.

What you will do

• Enable efficient parallelization on highly interconnected GPU clusters and design strategies for large-scale model training.
• Build capable solutions for multimodal LLM inference in high-throughput and low-latency scenarios.
• Profile, debug, and optimize GPU kernels and low-level systems configuration within our tech stack.
• Collaborate with researchers to ensure state-of-the-art models are implemented correctly and efficiently.
• Evaluate latest technologies and develop innovative ideas in computational possibilities.

Requirements

• Strong analytic skills and a scientific approach (Master's or doctoral degree in Computer Science, Math, Physics, or related field).
• Accomplished software engineer with a track record of project responsibility from conception to production.
• Strong experience in Python, including bindings to native C++ or Rust code.
• Deep familiarity with PyTorch and the AI software ecosystem (PyTorch Distributed, Megatron-LM, TorchTitan, TensorRT-LLM, vLLM, SGLang).
• Good understanding of GPU high-performance programming model and collective communication primitives (MPI or NCCL).
• English: High proficiency required.

Nice to have

• Experience optimizing and troubleshooting workloads on GPU compute clusters at scale.
• A thorough understanding of state-of-the-art Transformer architectures.

Culture & Benefits

• Diverse and internationally distributed team across 90+ nationalities.
• Open communication and regular feedback with an emphasis on empathy and growth mindset.
• Hybrid work schedule with flexible hours, requiring office presence twice a week.
• Virtual Shares, linking employee contribution directly to hirify.global’s growth.
• Regular in-person team events and monthly full-day hacking sessions.
• 30 days of annual leave and a competitive benefits package tailored to your location.