Senior AI R&D Engineer - Performance - Pre-training (f/m/d) arbeitnow Alephalpha Heidelberg · 10/24/2026

Primetime; Heidelberg; On-site; EUR 120,000 - 190,000; Full time

Benefits offered by this job

30 days of paid vacation

Wellhub fitness & wellness offerings

Subsidized company pension plan

Germany-wide transportation ticket

Budget for additional technical equip.

Flexible working hours and hybrid work

Job summary

Aleph Alpha in Heidelberg is seeking a Senior Performance Engineer to optimize pre-training efficiency for large-scale GPU clusters. You will profile training loops, tune parallelism, and remove bottlenecks to maximize throughput.

You will collaborate with researchers to define hardware-efficient model architectures and work hands-on on-site in Heidelberg to ensure high performance across thousands of NVIDIA GPUs. This role is part of Aleph Alpha Research GmbH's Training Efficiency Team.

Qualifications

• Proficient in Python and PyTorch
• Strong background in parallel/distributed systems
• Hands-on experience with ML techniques (LLMs)
• Deep understanding of CUDA
• Experience with distributed APIs (NCCL/MPI)
• Experience with profiling traces (PyTorch Profiler, Nsight)
• Role requires regular on-site collaboration in Heidelberg as part of Training Efficiency Team

Responsibilities

• End-to-End Optimization: Profile training loops with PyTorch Profiler, Nsight Systems and Nsight Compute to identify bottlenecks and maximize throughput.
• Distributed Strategy and Topology: Tune parallelism strategies (TP, DP, HSDP/FSDP, EP) for load balance and reduced critical-path bottlenecks.
• Hardware-Aware Modeling: Partner with researchers to define hardware-efficient model architectures without sacrificing convergence.

Skills

Python

PyTorch

Distributed systems

CUDA

NCCL

MPI

Profiling tools

Tools

Nsight Systems

Nsight Compute

PyTorch Profiler

Job description

Aleph Alpha is one of the few companies in Europe doing serious foundation model pre-training like our recently released Kolibri-1 model . Our customers - in finance, manufacturing, public administration - need models that understand German, meet European regulatory requirements, and work reliably in high-stakes settings. We're building that in Heidelberg.

We are hiring a Senior Performance Engineer to grow our pre-training efficiency team. If you are excited about making models fast and scaling them to thousands of GPUs, this is the role for you!

Team Culture

At Aleph Alpha, we foster a culture built on ownership, autonomy, and empowerment. Teams and individual contributors are trusted to take responsibility for their work and drive meaningful impact. We maintain a flat organizational structure with efficient, supportive management that enables quick decision‑making, open communication, and a strong sense of shared purpose.

About the role:

You will engineer the systems required to train foundation models at scale. Your objective is to maximize hardware utilization and training throughput on our large-scale GPU clusters (thousands of NVIDIA Blackwell GPUs). You will work at the intersection of deep learning frameworks, distributed systems, and GPU microarchitecture, eliminating bottlenecks from the Python layer down to the GPU kernel.

This role is for Aleph Alpha Research GmbH.

Your responsibilities

• End-to-End Optimization: Profile training loops using PyTorch Profiler, Nsight Systems and Nsight Compute to identify system- and kernel-level bottlenecks in order to maximize model throughput.
• Distributed Strategy and Topology: Configure and tune composite parallelism strategies (e.g. TP, DP, HSDP/FSDP, EP), optimizing load balance, minimizing critical-path bottlenecks, and managing communication-to-computation trade-offs for large-scale LLM training.
• Hardware-Aware Modeling: Partner with AI Researchers to define model architectures for hardware efficiency without compromising convergence.

Your Profile

Basic Qualifications

• Are proficient in Python and the PyTorch library.
• Have a strong engineering background in parallel and/or distributed systems with proven track record of excellence.
• Have hands-on experience with modern machine learning techniques (especially large language models and their life cycle).
• Deeply understand the CUDA programming model.
• Have experience in distributed programming with APIs like NCCL or MPI.
• Have experience analysing profiling traces with tools such as PyTorch Profiler and Nvidia Nsight.
• Please note this role requires regular on-site collaboration in Heidelberg as a member of the Training Efficiency Team.

Preferred Qualifications

• Contributions to modern distributed training frameworks (e.g., TorchTitan, Megatron-LM, DeepSpeed).
• Familiarity with low-precision training formats(MXFP4, MXFP8) and their impact on numerical stability and throughput.
• A deep understanding of NCCL communication primitives, NVSHMEM or CUDA IPC and their performance.
• A proven track record of implementing and optimising modern transformer-based model training.
• A proven track record working on the NVIDIA Blackwell architecture.

Compensation and Benefits

Become part of an AI revolution!

• 30 days of paid vacation
• Access to a variety of fitness & wellness offerings via Wellhub
• Substantially subsidized company pension plan for your future security
• Subsidized Germany-wide transportation ticket
• Budget for additional technical equipment
• Flexible working hours for better work‑life balance and hybrid working model

This overview was written from the original listing to help you understand the role. Always confirm the specifics on the employer’s application page.