StellenbeschreibungppWe are seeking a skilled engineer to join the Apertus post-training effort. The ideal candidate will develop, run, and evaluate the SFT and reinforcement learning pipelines used to turn Apertus base models into capable assistants. This role requires a strong background in LLM post-training, solid software engineering skills, and the ability to work collaboratively in a research-focused HPC environment. /ph3Project background /h3pWe train open foundation models with hundreds of billions of parameters on thousands of GPUs on one of the largest AI-ready supercomputers in Europe. The team counts more than a dozen full-time engineers working alongside leading researchers from EPFL and ETH Zürich, has released the Apertus 1 and Apertus 1.5 models, and works with over thirty academic collaborators to deliver fully open (open source), responsibly trained, multilingual, multimodal AI models for research and industry. /ppApertus is trained and developed on Alps, the Swiss National Supercomputing Centre's supercomputing infrastructure. The role requires someone who is comfortable working in an HPC environment and collaborating with researchers and infrastructure engineers. /ph3Job description /h3pThe engineer will contribute to the development, execution, and evaluation of scalable post-training workflows for Apertus. /ppInfrastructure and systems engineering /pulliBuild and maintain containerised environments for LLM post-training and RL workloads /liliAdapt containers and dependencies for execution on Alps / CSCS infrastructure /liliRun and monitor Slurm-based training and evaluation jobs /liliDebug failures related to distributed execution, checkpointing, filesystem performance, networking, and GPU utilisation /liliHelp maintain reproducible training recipes, configuration files, launch scripts, and documentation /liliWork with researchers and CSCS engineers to improve the reliability and performance of large-scale experiments /li /ulpLLM post-training and reinforcement learning /pulliSupport SFT, preference optimisation, and reinforcement learning workflows /liliBuild and run RL environments for tasks with verifiable outcomes, such as mathematics, code, tool-use, and reasoning /liliImplement and run reward modelling, reward calibration, and verifier-based training /liliGenerate and validate synthetic or gym training tasks /liliRun ablation studies comparing algorithms, reward functions, data mixtures, hyperparameters, and infrastructure settings /liliEvaluate model behaviour across reasoning, coding, mathematics, instruction-following, multilingual, tool-use, and safety benchmarks /liliDebug common post-training issues, including optimisation instability, reward hacking, regressions, and evaluation failures /li /ulh3Profile /h3pEssential /pulliMSc or PhD in Computer Science, Data Science, Artificial Intelligence, Machine Learning, or a related field /liliExceptional BSc candidates with strong engineering experience will also be considered /liliExperience in AI and neural network architectures /liliStrong collaboration and communication skills and ability to work across research and engineering teams /liliPrior hands-on experience in the core domains of this role is required /liliThis can be project or study based experience; formal work experience is preferred /liliA high degree of flexibility: priorities, tools, and day-to-day tasks shift with training schedules, releases, and a fast-moving field /liliHands-on experience with LLM post-training, be it alignment (SFT, preference optimisation) or reinforcement learning /liliThis means experience with frameworks such as veRL, slime, Megatron-LM, DeepSpeed, TRL, vLLM, SGLang, or similar tools /li /ulpStrongly preferred /pulliFamiliarity with distributed training concepts such as data parallelism, tensor parallelism, pipeline parallelism, checkpointing, and GPU communication /liliExperience with Slurm or another HPC workload manager /liliExperience building or adapting containers for HPC or GPU clusters /li /ulpNice to have /pulliPublished research in the domains relevant to this role, or familiarity with recently published research on these topics /liliExperience creating verifiable tasks for mathematics, code, reasoning, or tool use /liliFamiliarity with lower-level GPU/distributed libraries such as NCCL, Transformer Engine, FlashAttention, or communication backends /liliExperience with large-scale evaluation pipelines /li /ulh3We offer /h3ulliA stimulating academic environment at one of the world's leading technical universities /liliThe opportunity to work with state-of-the-art supercomputing infrastructure and cutting-edge AI research /liliCollaboration with top researchers and engineers from EPFL, ETH Zürich, CSCS, and other Swiss institutions /liliFlexible working arrangements, including options for remote work /liliProfessional development opportunities, including conference attendance and specialised training /liliThe chance to contribute to open-source projects with global impact /liliAccess to the broader Swiss academic ecosystem and industry partnerships /liliBeing part of Switzerland's sovereign AI development, working on technology with national significance /liliThe role can be based either in Lausanne at EPFL or in Zürich at ETH Zürich /li /ulh3We value diversity and sustainability /h3pIn line with our values, ETH Zurich encourages an inclusive culture. We promote equality of opportunity, value diversity and nurture a working and learning environment in which the rights and dignity of all our staff and students are respected. Visit our Equal Opportunities and Diversity website to find out how we ensure a fair and open environment that allows everyone to grow and flourish. Sustainability is a core value for us – we are consistently working towards a climate-neutral future. /p /p #J-18808-Ljbffr
