Build the
Future of
Immunotherapy

Responsibilities

• — Design, implement and train deep generative models (VAEs, transformers, and related architectures) on large-scale, high-dimensional data
• — Shape and constrain latent spaces (e.g. metric learning, contrastive objectives) so that learned latent spaces are controllable and support extrapolation to unseen states
• — Build count-aware decoders and likelihood heads (e.g. negative-binomial / zero-inflated parameterizations) and design composite training objectives with staged optimization
• — Construct careful data pipelines and samplers that prevent shortcut learning and control for confounders
• — Design rigorous benchmarks: zero-/few-shot generalization, ablations, and comparisons against state-of-the-art baselines
• — Translate model outputs into actionable predictions and rankings, and collaborate with domain experts to validate them

Required

• — MS or PhD in computer science, machine learning, statistics, physics, or a related quantitative field
• — Strong track record building and training deep neural networks in PyTorch
• — Solid grounding in generative modeling (VAEs, diffusion, flow-based, or autoregressive) and in modern transformer/attention architectures
• — Comfort with the math behind the methods: probability, linear algebra, optimization, and metric/representation learning
• — Experience working with large datasets and the engineering practices (efficient dataloading, distributed/GPU training, reproducibility) that make large-scale training tractable
• — Ability to design clean experiments, interpret results honestly, and communicate findings clearly

Preferred

• — Experience with single-cell genomics or other high-dimensional biological data such as scRNA-seq
• — Familiarity with computational biology tooling (e.g. Scanpy, AnnData, scVI) and single-cell foundation models
• — Publications or open-source contributions in ML or comp-bio

Responsibilities

• — Design and execute scRNA-seq experiments end-to-end, from cell preparation through library QC and sequencing handoff
• — Scale combinatorial barcoding protocols (Parse Evercode, SPLiT-seq, sci-RNA-seq) to 1M+ cells per run
• — Optimise protocols for cell fixation, nuclei isolation, and PBMC preparation across diverse sample types
• — Establish QC pipelines and enforce data quality standards across all experimental runs
• — Manage vendor relationships for reagents, consumables, and sequencing services
• — Collaborate with the ML team to translate experimental data into training-ready datasets
• — Recruit and mentor junior wet-lab staff as the team grows

Required

• — PhD in genomics, molecular biology, immunology, or a closely related field
• — Hands-on experience with combinatorial single-cell RNA-seq (Parse Evercode, SPLiT-seq, or sci-RNA-seq)
• — Proficiency in NGS library preparation and QC (TapeStation, Bioanalyzer, Qubit)
• — Experience with primary cell preparation: PBMCs, nuclei isolation, and cell fixation
• — Track record of troubleshooting and rescuing failed scRNA-seq runs

Preferred

• — Experience with Scanpy or Seurat for exploratory single-cell data analysis
• — Scripting ability in Python or R for QC automation
• — Background in oligo or barcode design for combinatorial indexing