> Markdown version of [/jobs/ext/4722810-senior-machine-learning-engineer-pre-training](https://www.wearedevelopers.com/jobs/ext/4722810-senior-machine-learning-engineer-pre-training). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Senior Machine Learning Engineer, Pre-training - **Company:** Boltz - **Location:** London, United Kingdom - **Experience:** Expert - **Contract:** Permanent contract - **Skills:** Clean Code Principles, Computer Clusters, Distributed Computing Environment, Python (Programming Language), Machine Learning, Node.Js, Open Source Technology, Management of Software Versions, Data Processing, Pytorch, Free and Open-Source Software, Machine Learning Operations, Data Pipelines - **Published:** November 15, 2026 - **Apply:** https://www.indeed.com/viewjob?jk=95ed8dc37bec27c7 ## About the Role * Senior or Staff-level experience building ML systems for large-scale training, with strong Python and PyTorch. * Experience with distributed training and data processing on multi-node GPU clusters. * Solid software engineering fundamentals: clean, tested, maintainable code and reproducible experiments. * A track record of shipping models or ML infrastructure that others relied on, such as a released model, a widely used codebase or a production system. * Close collaboration with research scientists, and enough ML intuition to judge how engineering choices affect results. Preferred Requirements * Experience with structure prediction, protein or molecular data, or diffusion models. * Experience maintaining open-source ML projects; early-stage startup experience. * Experience using Boltz models, or open-source contributions. We know great people don't always tick every box. If this role excites you and you meet most of the minimum requirements, we'd like to hear from you. ## Description As a Machine Learning Engineer on Pre-training, you will own the engineering behind our next open foundation model, Boltz-3: the data, training and evaluation systems that let us train it at scale and release it with state-of-the-art structure prediction. Every Boltz model starts here, and the open release is what 100,000+ scientists build on. You will work closely with the Pre-training team's research scientists, Research Engineering, who own training speed and cluster tooling, our Post-training team, and Platform Engineering. This role is ideal for an engineer who wants to build the systems behind a frontier open model and see it used across biotech and academia. What you'll do * Build and scale the data pipelines behind pre-training: ingesting, filtering, clustering and versioning large structural and sequence datasets. * Own the pre-training codebase and large training runs with research scientists, keeping experiments reproducible and runs healthy. * Build the evaluation and benchmark suites that show state-of-the-art antibody-antigen and small-molecule structure prediction. * Build inference infrastructure for large sampling runs, to measure how performance scales with inference-time compute. * Get each model ready for open-source release: packaging, documentation, inference code and community feedback that flows back to the team.