Full Stack Engineer

ABOUT SEVEN NODES

Seven Nodes is building a scientific reasoning platform for drug discovery. We bring together disease-specific omics, statistical modelling, biological knowledge graphs, mechanistic simulation, scientific literature and evidence provenance in one auditable workflow.

Our current product supports bulk RNA-seq workflows, from data ingestion and quality control through differential expression, pathway analysis, co-expression modules and Boolean network modelling. It produces ranked target hypotheses and reviewable target dossiers.

The platform is designed for in silico researchers and wet-lab experimentalists. Researchers can inspect how a conclusion was reached, trace evidence to its source, examine predicted network effects and decide what to test next. Our initial focus is oncology and immunology.

We are not building a generic chatbot or replacing scientific judgement. We are building software that makes target-selection reasoning explicit, reproducible and easier to challenge before a team commits time and capital to laboratory work.

ABOUT THE POSITION

We are looking for a Full Stack Engineer to build and own product engineering capabilities across the scientist-facing application, backend services, AI and ML integrations, and cloud workflows. You will work closely with the founders, computational biologists and other engineers to turn complex research workflows into dependable software.

This is not a conventional CRUD application. The product must handle large data uploads, long-running bioinformatics jobs, asynchronous state changes, scientific provenance, multi-step analyses and results that need to remain reproducible. The strongest candidate will be comfortable moving between interface decisions, API design, distributed workflows and production operations without losing sight of the scientific user.

You will not be expected to invent or validate biological models alone. The computational biology team owns scientific methods and interpretation. Your job is to make those methods usable, reliable, secure and inspectable in a production product.

RESPONSIBILITIES

• Own end-to-end product delivery. Take features from user problem and technical design through implementation, release, monitoring and iteration.
• Build the scientist-facing application. Create clear, responsive interfaces for data upload, analysis configuration, run status, results exploration, evidence review and target dossiers using TypeScript and React or an equivalent modern framework.
• Develop backend services and APIs. Build and maintain Python and FastAPI services, typed interfaces, validation, authentication, permissions and data access patterns that support reliable product workflows.
• Connect product flows to scientific computation. Integrate the application with containerised pipelines and long-running workloads using AWS services such as S3, Lambda, EventBridge, Batch, ECS and DynamoDB, together with Nextflow where appropriate.
• Productionise AI and ML capabilities. Build the application and service layers around model inference, LLM-assisted workflows and agent or tool interactions. Define structured inputs and outputs, evaluation checks, model and prompt versioning, fallbacks, source traceability and human review so the system remains useful when model behaviour is uncertain.
• Engineer reliable asynchronous workflows. Design for idempotency, retries, timeouts, cancellation, partial failure, progress reporting and recovery. Make system state understandable to both users and operators.
• Protect customer data. Apply secure defaults for uploads, storage, tenant separation, secrets, access control, auditability and data deletion. Contribute to the engineering controls needed for Seven Nodes' near-term SOC 2 Type I target.
• Make results traceable. Preserve the connection between inputs, parameters, pipeline versions, evidence sources, model outputs and the dossier a user reviews.
• Raise product quality. Write useful automated tests, improve CI and CD, add observability, investigate production issues and remove recurring failure modes.
• Work directly with scientific users. Translate research workflows into software without flattening scientific nuance. Ask precise questions, surface trade-offs and challenge requirements that do not solve the underlying problem.
• Improve the engineering system. Document decisions, simplify fragile areas and establish patterns that allow a small team to move quickly without creating avoidable operational risk.

WHAT SUCCESS LOOKS LIKE

ABOUT YOU

• Computer science education. You hold a Bachelor's, Master's or PhD in Computer Science.
• Production experience. You have at least 1-4 years of experience building and operating web applications used by real customers. Years are a guide; evidence of scope and judgement matters more.
• Strong front-end engineering. You are productive with TypeScript and React or a comparable modern framework. You can design component boundaries, manage complex client state and build interfaces for data-heavy workflows.
• Strong backend engineering. You can design and implement Python services and APIs, work with persistent data models and reason about concurrency, failure and backward compatibility.
• Applied AI and ML experience. You have shipped software that uses machine learning or generative AI in production. You understand inference and data flows, model or prompt versioning, evaluation, monitoring, structured outputs and the failure modes of probabilistic systems. Calling a model API without owning reliability and evaluation is not enough.
• Life sciences fluency. You have worked on software, data or AI systems used in life sciences or biotechnology. You can engage with researchers on provenance, reproducibility, scientific uncertainty and the difference between a computational result and validated biological evidence.
• Cloud fluency. You have deployed and operated systems on AWS or a comparable cloud and understand containers, event-driven systems, queues or job orchestration, storage, IAM and production diagnostics.
• Quality discipline. You use tests, code review, CI and observability to reduce uncertainty. You know which controls matter and avoid ceremony that adds little value.
• Security awareness. You understand common application and cloud risks and can apply sensible controls to customer data, authentication, authorisation, secrets and audit trails.
• Product judgement. You can turn an ambiguous user problem into a smaller, testable release. You care about whether a feature changes the user's decision, not simply whether the ticket is complete.
• Clear remote communication. You write decisions down, surface blockers early, give direct feedback and can work independently while remaining easy to collaborate with across India and the UK.

USEFUL BUT NOT REQUIRED

• Pharmaceutical or biotechnology R&D experience in target discovery, target validation, translational research, omics or preclinical workflows.
• Experience turning research methods into maintained products with computational biologists, data scientists or laboratory scientists.
• Hands-on work with LLM agents, retrieval-augmented generation, knowledge graphs, model evaluation or human-in-the-loop AI systems.
• Familiarity with Nextflow, AWS Batch, Docker and long-running compute workloads.
• Scientific visualisation, graph or network interfaces, large-file uploads or versioned analytical results.
• Multi-tenant B2B software, enterprise authentication, role-based access control or audit logging.
• Infrastructure as code, deployment automation, cost monitoring or SOC 2 readiness work.

HOW WE WORK

• Evidence before confidence. We separate what the data shows from what we infer and what still needs experimental validation.
• Small team, broad ownership. You will have room to shape the solution and will also be accountable for how it behaves in production.
• Direct communication. We discuss trade-offs openly, document decisions and expect people to challenge weak assumptions.
• Remote by design. The role is based anywhere in India. You must be able to work effectively from your location and maintain regular overlap with the London working day for collaboration.

COMPENSATION AND EMPLOYMENT

Compensation will reflect the scope of the role, relevant experience and the final employment arrangement in India. The complete package and employment terms will be discussed during the process and confirmed in writing before an offer is accepted.