Scientist / Senior Scientist, AI/ML

Closing Date: 01/04/2026

Location: Babraham Research Campus, Cambridge

Type: Full time, permanent / Start: Immediate

Salary: Competitive / Hours: 40 p/w

Hybrid - typically requiring at least one day per week in the office, with the understanding that employees may be required to attend more frequently if needed, or at their own preference.

Job Overview

At bit.bio, we are combining high-quality genetic perturbation data with AI and ML to advance computational approaches for deterministic programming. We are looking for a motivated and creative AI/ML Scientist to help unlock the value of our unique perturbation datasets in support of cell engineering. You will work with novel, large-scale perturbation datasets and apply cutting-edge computational methods to generate predictions and insights that inform experimental decision-making.

As an AI/ML Scientist, you will work across three connected areas: prioritisation of combinatorial perturbations, modelling cellular responses to genetic perturbation, and interpretation of perturbation-induced changes in cell state. You should be someone who can rapidly deploy and evaluate computational methods and work closely with computational and experimental teams to narrow combinatorial search spaces, generate robust predictions, and extract biologically meaningful insights that guide screening and discovery. Through this work, you will play a key role in establishing scalable computational capabilities that strengthen how bit.bio learns from genetic perturbation experiments.

Key Responsibilities

• Apply and adapt existing ML and AI methods for gene perturbation analysis, including training established perturbation models and fine-tuning pretrained foundation models on internal and external perturbation datasets.
• Establish robust, standardised evaluation workflows to benchmark perturbation-modelling performance on unseen perturbations and across new biological contexts.
• Deploy, evaluate and apply cutting-edge computational methods to prioritise combinatorial perturbations for experimental follow up and interpret perturbation induced changes in cell state.
• Work closely with computational and experimental teams to define modelling questions, refine datasets and metadata, and ensure computational outputs align with biological objectives.
• Contribute to best practices in computational analysis, model evaluation, and interpretation of large scale perturbation datasets.
• Keep abreast of advances in perturbation modelling, single cell analysis, and foundation models, and identify opportunities to apply emerging methods at bit.bio.

Qualifications

• PhD (or equivalent industry experience) in Computer Science, Machine Learning, Statistics, Computational Biology, or a related quantitative field.
• Experience developing and/or applying advanced AI/ML models in a research or industry setting to model gene perturbation responses using single-cell data.
• Comfortable working across machine learning, computational biology, and experimental science.
• Strong collaborator, used to working cross functionally in a fast moving research environment.
• Proactive, problem solving mindset and excellent written and verbal communication skills.
• Essential experience in training established perturbation models and/or fine tuning pretrained foundation models on novel biological datasets.
• Experience applying ML and AI methods to large scale single cell and gene perturbation datasets, including data preparation, model evaluation, and biological interpretation.
• Experience developing robust evaluation workflows for benchmarking perturbation modelling performance on unseen genetic perturbations and across new biological contexts.
• Python programming and modern ML frameworks such as PyTorch, JAX, or TensorFlow.
• Experience with computational approaches for identifying genes that drive cell state transitions, including gene regulatory network inference and related methods for perturbation target prioritisation.
• Experience with methods for cell type and cell state characterisation in single cell datasets, such as annotation, gene set scoring, pathway analysis, and related interpretive approaches.
• Experience working with multimodal datasets, such as paired transcriptomic, epigenomic, proteomic, or imaging data.
• Experience developing scalable ML workflows on cloud computing platforms such as GCP or AWS.
• Solid understanding of molecular and cellular biology concepts relevant to gene regulation, cell state, and perturbation response.
• Experience developing novel deep learning architectures and training foundation models for biological data.

Please apply via our careers page at