Biotech drug discovery platform integrating single-cell analysis, genomics, and ML
Relation pairs wet-lab biology (single-cell and multi-omics measurements from patient tissue) with machine learning and clinical data to identify and validate drug targets. The tech stack is modern infrastructure (Kubernetes, Terraform, ArgoCD) with heavy adoption of NVIDIA — pointing toward GPU-accelerated computational biology. Hiring velocity is accelerating across data (6 open roles) and research (5), while the active project list reveals they're scaling both ML pipelines and a hybrid on-prem/cloud environment, suggesting infrastructure constraints are limiting discovery speed.
Relation is a London-based biotech founded in 2019, developing medicines from patient-derived tissue using high-resolution biology and machine learning. The company operates an end-to-end model: extracting single-cell and multi-omics data directly from patient samples, running functional assays, applying ML to identify disease mechanisms, and advancing selected targets into drug development. Active programs include an osteoporosis therapy and a fibrosis collaboration with GSK. The 51–200-person team spans research, data science, engineering, and commercial functions, with recent hiring concentrated in data and research roles.
Kubernetes, Terraform, ArgoCD, GitHub Actions for infrastructure; Google Workspace and Microsoft 365 for collaboration. NVIDIA adoption signals GPU infrastructure for computational workloads. No legacy data warehouse stack detected.
Multi-omics data integration for drug discovery, target validation, ML training/inference pipelines, and hybrid cloud/on-prem infrastructure optimization. Active programs: osteoporosis development and GSK fibrosis collaboration.
Other companies in the same industry, closest in size