Senior Bioinformatics Scientist

Basecamp Research

London, United Kingdom

The Role

Our EDEN models, trained on the world's largest ethically-sourced biological dataset, have already designed novel antibiotics validated against priority pathogens in the lab, and prioritised vaccine targets faster than comparable genomic foundation models.

We are looking for a Senior Bioinformatics Scientist to join our Genomics team in London, working closely with our deep-learning modelling, engineering, and translational teams. You will focus on defining what data our models require, what “good data”means and validating through evidence the success of our foundational models.

You will be a first principles thinker capable of reasoning backwards: defining what data would train a capability, how to structure, and what evaluations would show the model is doing what we intend.

What you'll do

• Start from a modelling goal and work out the data behind it: what to assemble or curate, how to define ground truth and label quality, and how to build biologically sound splits homology- and structure-aware, not naive random so that what a model learns holds up.
• Design the evaluations that prove it: the metrics, baselines, and failure analysis that show a model is genuinely capable rather than exploiting leakage or a shortcut in the data.
• Fine-tune and rigorously evaluate genomic or protein models, or work hand in hand with those who do, so that your dataset decisions are grounded in how they actually move model performance.
• Reason about biological meaning from foundational, evolutionary-scale data through to human therapeutic context: target validation evidence, indication mapping, peptide-level questions.
• Run and develop the genomic and protein analyses that depend on sequence analysis, alignment, annotation, comparative and evolutionary approaches with the scepticism to spot when a result is an artefact of the method rather than the biology.
• Work reproducibly, so your datasets and analyses are something the team can rerun and build on rather than a result that only ran once on your machine using whatever tooling fits, from workflow managers to containers, without needing to be a platform engineer.
• Hold the line on data integrity, treating provenance and lineage as first-class parts of the dataset because a model is only as trustworthy as the consented, traceable data underneath it.

About You

• You have demonstrated experience curating data specifically for life-science AI, so you understand what separates a dataset that trains a good model from one that quietly poisons it. That comes from having been close to the models themselves. You've trained, fine-tuned, or run inference on them enough to know what a dataset looks like from the other side.
• Your foundations are strong. You have core bioinformatics, sequence informatics and the standard genomic and protein toolchain, along with the judgment to know where those tools quietly break down, backed by sound software practice in reproducibility, version control, and pipelines.
• You have direct experience in human research, functional genomics, or clinically relevant work, the kind where the biology connects to human health rather than sitting at arm's length.
• You'll likely have a PhD in a relevant field, an MSc with 1-3 years of relevant experience, or a BSc with 3 or more years.
• Low ego, collaborative instincts, and a startup mentality. You're comfortable with ambiguity and happy to wear multiple hats in a team where everyone contributes towards the company's success.