> Markdown version of [/jobs/ext/3515987-machine-learning-research-engineer](https://www.wearedevelopers.com/jobs/ext/3515987-machine-learning-research-engineer). 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). --- # Machine Learning Research Engineer - **Company:** Seqera - **Location:** Barcelona, Spain (Remote available) - **Contract:** Permanent contract - **Skills:** Application Programming Interfaces (APIs), Computational Biology, Python (Programming Language), Machine Learning, Language Modeling, Open Source Technology, Workflow Management Systems, Pytorch, Build Management - **Published:** September 22, 2026 - **Apply:** https://www.indeed.com/viewjob?jk=84362803b94ad3f3 ## About the Role * You have post-trained open-weight language models - SFT and at least one preference-tuning method - and put the results in front of real users, not just in a notebook., * Computational biology, chemistry or pharma exposure. * Familiarity with Nextflow or workflow orchestration. * Experience evaluating agentic, multi-step, tool-using systems. * Distillation or small-model specialisation from frontier models. * Publications or technical reports in post-training or continual learning. ## Description We are hiring a Research Engineer to own the improvement loop for a research engine we are building on top of the Seqera Platform. You will build the data and evaluation flywheel from engine' own run data, and use it to post-train small, task-specific models that make it better at generating, ranking and running scientific ideas. You will work directly with our Founding Scientist and the engineers who own the application, on a small team that moves fast and ships to real scientists. This is an in-office role in Barcelona. The team is here, the whiteboards are here, and this is the kind of work that goes faster in the same room., * Design and build the evaluation suite that defines what "better" means for an autonomous research agent. * Build the pipeline that turns messy, real-world run data from deployments into a training-ready corpus. * Post-train small open-weight models - SFT, preference tuning, adapters - to beat frontier-API baselines on core tasks at lower cost and latency. * Run repeated training cycles on evolving data. * Participate in journal clubs, stay at the frontier of post-training specialized models. * Own the Next-Gen roadmap with our Founding Scientist: what data unlocks what capability, and in what order., * You have retrained models on changing data more than once and have dealt with regression and forgetting in practice: replay, data mixing, adapter strategies, eval gating. * You have built evaluation harnesses for a specific task, defined the metric yourself, and defended the result to someone sceptical. * You work fluently in Python with the modern stack - PyTorch, transformers, TRL, PEFT, vLLM or their equivalents. * Your open-source footprint speaks for you: substantive contributions to the tools above, or fine-tuned models on the Hub. * You are an operator. When the pipeline you need does not exist, you build it. You are comfortable when the data, the tooling and the direction are all incomplete at the same time. * You want to work in-office in Barcelona, or you are ready to move here. ## Related Videos - [Geometric deep learning for drug discovery](https://www.wearedevelopers.com/videos/264-geometric-deep-learning-for-drug-discovery) - [Introduction to Azure Machine Learning](https://www.wearedevelopers.com/videos/368-introduction-to-azure-machine-learning) - [Photonic Computing: Programming a New Class of AI Accelerators (incl. Live Coding)](https://www.wearedevelopers.com/videos/100196-photonic-computing-programming-a-new-class-of-ai-accelerators-incl-live-coding) - [Answering the Million Dollar Question: Why did I Break Production?](https://www.wearedevelopers.com/videos/1171-answering-the-million-dollar-question-why-did-i-break-production) - [Serverless deployment of (large) NLP models ](https://www.wearedevelopers.com/videos/158-serverless-deployment-of-large-nlp-models) - [What non-automotive Machine Learning projects can learn from automotive Machine Learning projects](https://www.wearedevelopers.com/videos/397-what-non-automotive-machine-learning-projects-can-learn-from-automotive-machine-learning-projects) ## Related Articles - [What Are Large Language Models?](https://www.wearedevelopers.com/magazine/304-what-are-large-language-models) - [MLops – Deploying, Maintaining And Evolving Machine Learning Models in Production](https://www.wearedevelopers.com/magazine/115-mlops-deploying-maintaining-and-evolving-machine-learning-models-in-production) - [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer) - [Why Upskilling And Reskilling is Important For Developers](https://www.wearedevelopers.com/magazine/428-why-upskilling-and-reskilling-is-important-for-developers) - [MLOps – What’s the deal behind it?](https://www.wearedevelopers.com/magazine/125-mlops-what-s-the-deal-behind-it) - [Fully Remote Software Engineer Jobs](https://www.wearedevelopers.com/magazine/447-fully-remote-software-engineer-jobs)