Full-Time
Platform Engineer
Relation Therapeutics
51-200 employees
Drug discovery platform using ActiveGraph ML
No salary listed
London, UK
Hybrid
Hybrid work in London is required.
Bachelor's
 DevOps & Infrastructure (1) |
|---|
Get referred to Relation Therapeutics
See people who can refer or advise you
- A Bachelor's degree in Computer Science or a related quantitative field.
- At least 3 years of combined experience in software engineering, including the ability to design, deploy, and manage cloud infrastructure and Kubernetes clusters.
- Experience setting up and maintaining continuous integration and continuous delivery pipelines using tools such as GitHub Actions and ArgoCD.
- Experience with infrastructure as code tools such as Terraform.
- Solid Python coding experience and the ability to write clean, efficient, and maintainable code.
- Strong analytical and troubleshooting skills with a systematic approach to solving complex problems.
- Strong verbal and written communication skills to collaborate effectively with cross-functional teams.
- Develop, evolve, and contribute to the architectural strategy for Kubernetes-based clusters across private and multiple public clouds.
- Engineer and manage the full lifecycle of internal and public-facing applications, APIs, and services.
- Enhance continuous integration and continuous delivery pipelines to facilitate efficient development, testing, and deployment processes.
- Establish comprehensive monitoring solutions to proactively detect and resolve issues.
- Ensure robust backup, disaster recovery, and failover strategies are in place.
- Enforce best practices and security measures to protect sensitive data and comply with industry standards.
- Work closely with Machine Learning and Data Science teams to understand their computational needs and challenges, and provide expert guidance and support.
- Experience developing and deploying applications on Kubernetes.
- Experience with Go.
- Experience with DataOps and MLOps.
Relation Therapeutics uses machine learning to speed up drug discovery by combining a large-scale ActiveGraph ML platform with a Lab-in-the-Loop approach. Its system analyzes how genes, proteins, and drugs relate to each other to understand biology and predict promising therapeutics, using real cells from proprietary biobanks to generate genomic data that continuously informs new experiments. It differs from others by applying ActiveGraph ML at a large scale in drug discovery and integrating active learning with experimental validation in a closed loop. Its goal is to discover treatments for currently untreatable diseases and license successful candidates to pharmaceutical companies for development and commercialization.
Company Size
51-200
Company Stage
Late Stage VC
Total Funding
$139M
Headquarters
London, United Kingdom
Founded
2019
Get referred to Relation Therapeutics
See people who can refer or advise you
Simplify's Take
What believers are saying
- GSK expanded its collaboration on July 30, 2026, to $110 million.
- Novartis signed Relation in December 2025 with $55 million upfront and $1.7 billion milestones.
- Hiring remained active through July 2026, signaling expansion instead of layoffs.
What critics are saying
- MORGAN needs massive wet-lab throughput; any data-quality failure breaks the entire platform thesis.
- Relation still sells preclinical targets, so a single failed Novartis or GSK program hits credibility.
- Big pharma can internalize perturbation workflows by 2027, shrinking Relation’s edge and pricing power.
What makes Relation Therapeutics unique
- MORGAN trains on petascale human perturbation data, not static association datasets, since July 2026.
- Relation pairs Lab-in-the-Loop automation with human tissue biology, compressing target discovery feedback cycles.
- GSK and Novartis back Relation’s causal biology platform, validating its upstream target-ID strategy.
Help us improve and share your feedback! Did you find this helpful?
Benefits
Hybrid Work Options
Growth & Insights and Company News
6 month growth
↑ 4%1 year growth
↑ 2%2 year growth
↑ 9%GSK will establish a new global R&D centre on the Cambridge Biomedical Campus in the UK, accommodating over 1,000 scientists. The facility will support closer collaboration with nearby research institutions and biotech companies. The pharmaceutical firm has also agreed a research collaboration with Relation Therapeutics. The partnership will use advanced cellular biology models and large-scale data generation. The moves signal GSK's increased focus on data-driven and AI-supported drug discovery. The Cambridge location places the company at the centre of one of Europe's leading life sciences hubs. GSK shares currently trade at £19.32, approximately 10% below the £21.42 analyst consensus target.
Relation builds causal AI for drug discovery. New foundation model and an expanded GSK partnership reflect a growing shift from observational biology towards mechanistic understanding. Artificial intelligence has become an increasingly familiar fixture in drug discovery, but not all biological models are asking the same question. Many seek patterns within vast observational datasets - identifying associations between genes, cells and disease. Relation Therapeutics believes the next leap requires something different: models trained not simply to recognize what changes alongside disease, but to understand what happens when biology is deliberately perturbed. That philosophy sits behind two announcements from the London-based biotechnology company. Relation has unveiled MORGAN (Multi-Omic Regulatory Genomics using Artificial Neural Networks), a foundation model trained on large-scale perturbation datasets generated from human cellular systems [1], while simultaneously expanding its strategic collaboration with GSK to deepen the application of its Lab-in-the-Loop platform across therapeutic discovery [2]. Together, the announcements reflect the company's ambition to couple machine learning with experimental biology in a continuous cycle, allowing computational predictions to inform laboratory experiments - and laboratory results to refine the model. Longevity.Technology: Longevity has become remarkably adept at describing aging. Longevity Ltd. can estimate biological age, stratify populations and identify molecular signatures associated with healthspan. Yet describing biology and understanding biology are not the same thing. Much of today's aging biomarker landscape remains fundamentally correlational - enormously valuable for prediction, largely silent on the question that actually matters: what should Longevity Ltd. perturb to change the outcome? Relation's latest platform is interesting not because it claims to be a longevity technology, but because it is built around the sort of causal, interventional framework that geroscience increasingly needs. As the field matures, the winners may not be those who collect the largest datasets, but those who can tell association from cause. Adjacent, not identical. Relation's therapeutic focus - immunology, metabolic disease, bone disorders - has nothing to do with aging, officially. Unofficially, all three diseases get considerably more common the longer a person is alive, which is the sort of overlap no biotech can quite dodge. Asked whether MORGAN is being developed with aging biology in mind, CEO David Roblin is careful not to overstate the connection. "Many of the diseases we focus on become more prevalent with age, so there is naturally an overlap between disease biology and aspects of aging biology," he told Longevity.Technology. "Our primary objective, however, is to better understand the cellular mechanisms that drive human disease and use that knowledge to discover new therapeutic targets." That distinction matters. Rather than training MORGAN to identify age-related signatures specifically, Relation is building what Roblin describes as a broad biological foundation capable of interpreting cellular responses across multiple contexts. "MORGAN is designed to learn how human cells respond to genetic and pharmacological perturbations across different biological contexts," he explains. "By generating large-scale, high-quality multi-omic perturbation datasets, we're building AI models like MORGAN that can distinguish meaningful biological states and responses." As the model expands to incorporate increasingly diverse datasets, he believes it may eventually provide deeper insights into "how different factors - including age, disease and therapeutic intervention - shape cellular behavior." Cause and effect, not just correlation. Perhaps the more intriguing aspect of Relation's approach is not the model itself, but the data upon which it is trained. Much of contemporary aging research still runs on longitudinal cohorts, biobanks and biological clocks - observational data, essentially, however sophisticated the maths dressing it up. It has transformed what the field can measure: age, trajectory, risk, candidate biomarkers by the thousand. What it cannot do, however carefully it is stratified, is tell you what to do about any of it. "Our approach starts with understanding causal biology through controlled perturbation experiments rather than observation alone," Roblin says. "Longitudinal cohorts and biobanks tell us what is associated with disease, while perturbation data helps us understand cause and effect - how cells respond when specific pathways are altered. This is essential to get at the biology causing disease and therapeutics which can reverse. We believe those two approaches are highly complementary." Rather than presenting perturbation biology as a replacement for cohort science, Relation sees the approaches as addressing different questions - one descriptive, the other mechanistic. Foundations first. For geroscientists, an obvious question follows. Could a platform built on causal cellular biology ultimately contribute to aging biomarkers or even the discovery of geroprotective interventions? Roblin does not dismiss the possibility, but neither does he suggest it is the company's immediate objective. "Today, MORGAN is being developed to improve target discovery and therapeutic development across disease areas by building predictive models of cellular behavior," he says. "As the platform grows, we expect it to generate insights that extend beyond individual indications because many biological pathways are shared across diseases and physiological processes." He adds that whether those insights eventually contribute to aging biomarkers or therapies targeting the biology of aging remains "an exciting scientific question," while emphasizing that Relation's near-term priority is "building robust foundation models that deepen our understanding of human biology and enable the discovery of better medicines." That emphasis on infrastructure rather than indication also appears to underpin the expanded GSK collaboration. Rather than focusing on a single target, the partnership continues to integrate human data, wet-lab experimentation and machine learning to improve target identification across complex diseases. Mechanisms matter. The excitement surrounding AI in biomedicine often centers on scale - larger models, larger datasets and greater computational power. Relation's announcements hint at a different trajectory. If the next generation of biological AI is judged less by how accurately it predicts associations and more by how effectively it identifies mechanisms, perturbation biology may prove just as significant as model architecture. For longevity research, where understanding why biology changes increasingly matters as much as measuring that it has changed, that distinction could become increasingly difficult to ignore. Photographs courtesy of Relation Therapeutics. LTUI: 2372 Relation Therapeutics. Powered by: Longevity level(s): * Level 6: Aging disease prevention therapeutics Overview. Relation Therapeutics is a clinical-stage biotechnology company headquartered in London that integrates single-cell multi-omics, functional assays, and machine learning to discover transformational medicines. The company employs a Lab-in-the-Loop platform combining patient-derived tissue data with computational systems to identify novel drug targets across multiple disease areas, with initial focus on bone-related diseases, fibrotic conditions, and atopic diseases. Relation's approach aims to translate high-resolution biological understanding into therapeutics that address some of the most challenging human diseases by linking genetic cause to clinical phenotype. Access all data points on DLT: * Financial Data * Patents * Technology * Management * Advisory Board * News * Classifications * Similar Companies
Relation has announced MORGAN, a cellular foundation model designed to predict how human cells respond to genetic and pharmacological interventions. The model aims to provide insight into disease mechanisms and help identify therapeutic opportunities. MORGAN combines large-scale computation with multi-omic perturbation data generated in automated laboratories. The company will create petascale datasets using high-throughput cellular perturbation experiments designed for consistency beyond conventional laboratory approaches. Each tissue-specific MORGAN model focuses on a therapeutically important cell type central to disease biology. The general-purpose model can be applied across different cell types and disease areas. Chief executive David Roblin said the platform represents infrastructure needed to accelerate medicine discovery by combining frontier AI with industrial-scale data generation.
AI is transforming drug design - But not the bottleneck that matters most. Over the past five years, the narrative around AI in drug discovery has centred on a couple of breakthroughs: The first was AlphaFold. What started as a revolution in protein structure prediction has quickly expanded into predicting the shapes of RNA, antibodies, complexes, and entirely new biological assemblies as seen in the more recent Alphafold 3 model. This shift has spawned an ecosystem of companies racing to design better molecules by exploiting improved 3D structural insight. The other development was of course the advent of the large language model as a serious piece of kit with GPT-3 and chatGPT. It enabled researchers to imagine what would be possible when most problems in compound design Machine learning could be reframed using these tools: How can Life Sciences Week design a medicine to better target a protein of interest based on the fact that Life Sciences Week know the shape of its compound and protein? Life Sciences Week has witnessed extraordinary achievements, but there now may be a system-level imbalance in where investment is going based on what Life Sciences Week can do now versus what is the main pain point and resource sink in drug development. In the last 5 years a striking proportion of biotech funding has flowed into platforms building better binders: generative chemistry models, structural docking, de novo protein design, biologics modelling, and molecular shape optimisation. However, there are signs the tide is beginning to turn. Over the last 18-24 months, several new platforms and partnerships have explicitly focused on AI-driven Target-ID rather than downstream chemistry. Launches such as Genomics plc's Mystra, new clinically anchored Target-ID platforms from Owkin, and partnerships like AstraZeneca-Tempus-Pathos and Exscientia-Sanofi point to a shift in emphasis toward upstream biological decision-making. Even in the last month of 2025 Life Sciences Week saw new partnerships with an eventual value of around $1.7B for Relation Therapeutics, a British AI drug discovery using a lab-in-the-loop approach, with Novartis, and Deerfield Management. While design-focused AI still dominates total capital, Target-ID is clearly emerging as a growing and increasingly investible category. Most of the commercial AI-drug discovery platforms seem to sit firmly in the drug design space where the investment story is simple "Here is a better, cleaner molecule we designed with AI". Perhaps as shareholders and investors have become more aware of the true bottlenecks in drug development - the investment landscape is changing. The problem is that the drug design bottleneck is not what breaks R&D productivity. The biggest failure point is efficacy - Choosing the wrong target. For a decade, Life Sciences Week has known that the largest and most expensive source of attrition in R&D is Phase 2 clinical failure. Depending on the therapeutic area, roughly half of all Phase 2 trials fail on efficacy alone. By this point a company may have spent $50M+ building and testing a candidate that never had a chance. What does a phase 2 clinical trial really measure? Phase 2 clinical trials are the first point where Life Sciences Week test whether a drug actually works in patients with the disease. They assess efficacy, not just safety. These trials are large, expensive, and where most drug candidates ultimately fail. Improving Phase 2 outcomes depends heavily on choosing the right biological target at the very start of the pipeline. Peer-reviewed analyses published over the last decade confirm the scale of this problem. Analyses of 2010-2017 clinical trial data estimate that around 40-50% of clinical development failures are driven by lack of clinical efficacy (Sun et al., 2022). In Phase II specifically, cross-industry and company-level datasets suggest that roughly 50-70% of Phase II terminations are primarily due to insufficient efficacy (Arrowsmith & Miller, 2013; Feijoo et al., 2020; Wu et al., 2021). Late-stage analyses of pivotal trials show a similar pattern, with 57% of failed agents in Phase III or at registration failing for lack of efficacy (Hwang et al., 2016), and some therapeutic areas such as neurology and respiratory disease seeing more than 70-80% of failures attributed to poor efficacy rather than safety. Together, these findings emphasise that the biological hypothesis behind the drug, not the chemistry, breaks most often. Against this backdrop, human genetics offers one of the strongest forms of target validation. Nelson et al. (2015) demonstrated that drug programmes backed by human genetic evidence are twice as likely to succeed from Phase II onwards. More recent large-scale analyses show 2-3x higher likelihood of success when the chosen target has robust genetic support (Minikel et al., 2024, King et al., 2019). Life Sciences Week has been optimising the "wrong" Part of the pipeline. Drug discovery is like archery. Life Sciences Week has spent billions optimising the arrows - making them fly straighter, faster, and more precisely. Life Sciences Week has even upgraded the bow with generative models and multimodal AI. But if Life Sciences Week is aiming at the wrong target, none of that matters. Better arrows don't compensate for poor aim. Choosing the right biological target is the act of aiming, and it determines whether all the downstream engineering can succeed. AI-accelerated drug design, without better target selection and trial design, increases the rate at which programmes reach the most expensive point of failure. The industry as a whole seems to be learning this. Why has Target Identification had comparatively Little AI Attention until now? 1. Data scarcity: While chemical and structural data is abundant, high-confidence causal datasets linking proteins to disease are rare and often closely guarded. 2. Biological complexity: Targets act within networks. Causal inference is hard, especially in multifactorial diseases, although there are some techniques that can help disentangle causality. 3. Delayed feedback loops: You don't know if a target works until years after a clinical trial begins. However, this doesn't mean Life Sciences Week should just accept the situation. Life Sciences Week can provide information to calibrate the "Big Bets" that companies make at these early stages. 4. Commercial optics: It's easier to pitch "AI-designed molecules" than probabilistic target validation. Perhaps shareholders want to see the organisation adopting AI. Drug design, quite rightly, is the first place to try to use it. As a result, while investment into generative AI for chemistry is booming, AI applications for target identification receive a fraction of the funding, although that is increasing. Historically, a substantial majority of startup and partnership capital in AI-enabled drug discovery has flowed into generative chemistry, protein design, and modality optimisation, with only a smaller share directed toward companies working on Target-ID or efficacy prediction. While exact proportions vary across analyses and years, most investment tracking reports agree that funding has been significantly weighted toward design-focused platforms, even as a growing number of investors and biopharma partners are beginning to prioritise upstream biological decision-making. This leaves the industry over-investing in downstream engineering and under-investing in the upstream biological decisions that determine whether a drug will work.
Mishcon de Reya advises Relation Therapeutics on strategic multi-programme collaboration with Novartis. Mishcon de Reya has advised Relation Therapeutics on its multi-programme, strategic collaboration with Novartis to discover and advance novel targets for atopic diseases. Under the terms of the agreement, Relation will receive $55 million comprising an upfront payment, equity investment, and additional R&D funding. In addition, Relation is eligible to receive preclinical, development, regulatory and commercial milestones of up to $1.7 billion, along with tiered royalties on new sales of products. The collaboration pairs Relation's AI-powered drug discovery platform and human data generation capabilities with Novartis's deep expertise in immuno-dermatology to identify, validate, and advance potential first-in-class targets in atopic diseases driven by immune dysregulation. Benjamin Swerner, COO and founder of Relation, said: "This is Relation's third major collaboration in 12 months and the third time we have engaged Patrick, David and the Mishcon life sciences group to advise us. They are embedded in our team and understand our technology and business model. We get from Patrick and his team seasoned experience, and pragmatic legal advice in relation to the complex issues that arise." Patrick Farrant, Head of Life Sciences at Mishcon de Reya, commented: "Relation are an amazing business and a wonderful group of people to work with. We are truly privileged to support them on this latest transaction. Relation's platform using AI and human data to deliver targets that meet the needs of patients, is transformative and at the forefront of tech bio." The Mishcon de Reya team was led by Patrick Farrant alongside Dr David Rainford and Sophie Wood. The equity investment was led by Sarah Palmer of Mintz, Levin, Cohn, Ferris, Glovsky and Popeo, P.C.