AI Engineer - Large Scale Pre-Training

Vesoma; München; On-site; EUR 90,000 - 130,000; Full time

Job summary

Vesoma in Munich, Germany, is hiring an AI Engineer for Large-Scale Pre-Training to architect scalable training infrastructure for multi-billion-parameter robot foundation models. You will work with AI leadership and researchers to optimize data pipelines and experiment management for high-throughput GPU clusters.

The role requires 3+ years of hands-on experience with distributed training, core stack proficiency, and a startup-minded, rigorous engineering ethos suitable for a global industrial

Qualifications

• 3+ years hands-on experience training multi-billion-parameter models on multi-node GPU clusters.
• Experience supervising long-running experiments, in-flight debugging and optimizing goodput.
• Strong expertise in distributed training and training efficiency at scale.

Responsibilities

• Scalable Training Infrastructure: Develop software infrastructure to train multi-billion-parameter models on multi-node GPU clusters.
• Data Pipelines: Build high-throughput data pipelines that stream, decode and preprocess petabyte-scale video datasets, keeping GPUs fully utilized.
• Experiment Management: Build tools to organize, monitor and optimize large training workloads and model sweeps.

Skills

Distributed Training

Training Efficiency

Fault Tolerance

Core Stack

Mindset

Tools

Megatron-Core

DeepSpeed

TorchTitan

JAX/GSPMD

Slurm

Kubernetes

PyTorch

CUDA

NCCL

Triton

NVDEC

WebDataset

Mosaic Streaming

Ray Data

DALI

Job description

As an AI Engineer for Large-Scale Pre-Training, you enter as a key member of the early AI team. You will work directly with our AI leadership and researchers to define the infrastructure we use to train large-scale robot foundation models on petabyte-scale datasets across our GPU fleet. This infrastructure is a critical part of our research and development cycle: it must strike the right balance between stability and velocity, enabling fast research iterations based on reliable results. This is a full-time, permanent position.

Your profile

Key responsibilities

• Scalable Training Infrastructure: Develop software infrastructure to train multi-billion-parameter robot foundation models efficiently on large video datasets across multi-node GPU clusters.
• Data Pipelines: Build high-throughput, reliable data pipelines that stream, decode and preprocess petabyte-scale video datasets, keeping our GPUs fully utilized and supporting deterministic, resumable training.
• Experiment Management: Build the tools to organize, monitor and optimize large training workloads and model sweeps. Remove bottlenecks in our training pipelines and make sure all our runs maintain the highest possible goodput.

Your profile

We seek people who embrace the empirical nature of AI research and want to accelerate our research velocity with best-in-class tools - so that we can all make breakthroughs faster, together.

• Experience: 3+ years of hands-on experience training multi-billion-parameter models on multi-node GPU clusters, including supervising long-running experiments, in-flight debugging and optimizing goodput and MFU.
• Core Skills: Deep expertise in distributed training and training efficiency at scale. In detail:
• Distributed Training: Data parallelism and sharding (DDP, FSDP/FSDP2, ZeRO), tensor, pipeline and sequence parallelism, and frameworks such as Megatron-Core, DeepSpeed, TorchTitan or JAX/GSPMD.
• Training Efficiency: BF16/FP8 mixed precision, activation checkpointing, FlashAttention, and profiling with PyTorch Profiler or Nsight Systems.
• Fault Tolerance: Distributed and asynchronous checkpointing, automatic recovery, and diagnosing NCCL and hardware failures.
• Core Stack: Python, PyTorch and/or JAX, CUDA, NCCL, and cluster orchestration with Slurm or Kubernetes.
• Mindset: You balance the agility of a startup with the professional rigor required for a global industrial player. You are a curious builder who thrives in a collaborative, fast-paced research environment, and an engineer with a research mindset who understands the workflows of cutting‑edge AI research and supports them with rigorous software engineering.
• A plus: High-throughput data pipelines for large video datasets (WebDataset, Mosaic Streaming, Ray Data, DALI/NVDEC); custom kernels in Triton or CUDA; familiarity with multimodal or robot foundation models (VLAs, video transformers, diffusion/flow‑matching policies); experiment tracking and sweep infrastructure (Weights & Biases, MLflow, Hydra).

About us

We are building a global leader in physical AI, anchored in Europe’s strongest robotics heritage. Founded by entrepreneurs with backgrounds in scaling digital products and advanced robotics, the team is bolstered by senior leadership from top-tier AI research labs and industrial tech companies.

Driven by a mission to create versatile and trustworthy embodied AI for physical labor, We fuse deep engineering mastery in AI and mechatronics with a rapid-execution culture. We are addressing the labor shortage by closing the gap to production‑grade deployment. This means building the trusted alternative: robots that are safe, secure, and engineered to automotive standards of scale and durability.

Our Locations

We have two main locations: Munich, Germany and Limassol, Cyprus. This pan-European setup represents our core ambition of building a European technology leader, gives us a great talent base and a strong corporate structure.

We are an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all builders.