AI Research Engineer, Safety Pretraining for Open Frontier Models, EPFL (x3)

Research topics:

• AI safety & alignment
• Large language models
• Pretraining at scale
• Synthetic-data generation
• Model evaluation & red-teaming

We are hiring three AI research engineers to build the safety-pretraining stack for the next generation of Apertus, one of the world’s largest fully-open LLM programs and the largest pretraining effort in the non-commercial space.

The positions are hosted jointly by

• Swiss AI Initiative,
• Data Science & AI Lab (dlab, Prof. Robert West),
• Machine Learning & Optimization Lab (MLO, Prof. Martin Jaggi).

Most alignment today is postprocessing tacked onto a model whose representations have already crystallized during pretraining—what we call “lipstick-on-a-pig alignment”. We take a different view. We raise models: value formation is woven throughout the entire training process, starting from token 0 in pretraining, so that alignment is bound up with capability rather than layered on top of it (high-level vision here). Our early results are encouraging.

As an AI engineer on this project, you’ll translate these methods into Apertus’s production LLM pipeline scaling to thousands of GPUs, designing and applying production recipes for a new 700B model trained from the ground up starting in 2026.

The three roles

We are hiring for three complementary positions. You can apply for one or several.

• Training systems. Own the Megatron-LM integration: distributed training across thousands of GPUs, throughput optimization, scaling from 1B to 700B parameters (both MoE and dense models). You keep large runs fast and healthy.
• Data infrastructure. Own the synthetic-data pipeline: teacher-LLM orchestration, data generation and quality control, and the adaptive-curriculum machinery that decides when a model is ready to learn from a given example.
• Evaluation & red-teaming. Own the evaluation harness: inference optimization, LLM-as-a-judge infrastructure, reproducibility, ablation hygiene, automated jailbreak and persona-drift testing, and interpretability tooling that tracks what’s happening inside models across checkpoints.

What we offer

• Serious compute. Secured access to the Swiss National Supercomputing Centre’s Alps cluster (10,000+ GH200 GPUs), with a budget exceeding 10M GPU-hours per year for pretraining.
• Real production impact. Methods that pass validation feed directly into the next Apertus training run. You’ll build a real production LLM, not merely a prototype that might one day matter.
• Highly competitive compensation, well above standard academic scales.
• Paper co-authorship. While your focus is on engineering, you’ll also contribute as a co-author to the research papers that the team will publish.
• An open, international research environment, an extremely well-funded national research system, generous travel support, and an office next to a stunning lake and even more stunning mountains.

Qualifications

• A Master’s degree (or equivalent experience) in computer science, engineering, or a related field. A PhD is not required.
• Relevant engineering experience with deep learning in general, and LLMs in particular. Having shipped systems, not only prototypes, is a plus. Experience with training and handling large MoE models is a plus.
• Strong programming skills and hands-on experience with modern ML frameworks. Depending on the role: distributed/large-scale training (Megatron-LM, PyTorch, GPU performance), data pipelines and LLM orchestration, or evaluation/red-teaming and interpretability tooling.
• A pragmatic, quality-obsessed engineering sensibility, and genuine interest in the mission of safe, open AI.

Time frame

Positions are for two years, with an ideal start in 2026. There is some flexibility on start date.

About EPFL

EPFL ranks among the world’s top universities in computer science. It sits in Lausanne, Switzerland, a vibrant, highly international city in an Alpine setting on the shores of Lake Geneva, in the heart of Europe. English is the main language spoken at EPFL, and no French is required.

How to apply

To apply, please complete this short online form.

Review of applications begins immediately and continues until the positions are filled.

Contact:

• Robert West: robert.west@epfl.ch · dlab.epfl.ch/people/west
• Martin Jaggi: martin.jaggi@epfl.ch · people.epfl.ch/martin.jaggi