Senior Principal Forward Deployed DevSecOps Engineer

We are looking for a Senior Principal Forward Deployed DevSecOps Engineer to work closely with scientists and bring AI, automation, and data into the lab.

As a Senior Principal Forward Deployed DevSecOps Engineer in Labs & CMC Digital & Tech, you will work directly with our Product Owners, scientists, and data engineering teams to shape and deliver digital products for the lab.

Key Responsibilities

• Shaping the solution: working with Product Owners and scientists to understand the problem, pairing with them on the options, and driving towards a solution that is technically sound and fits how the lab works.
• Moving quickly: designing and building PoCs, PoVs, and MVPs on real lab data in days or weeks, and giving a clear view on whether each one should go further.
• Automating the lab: connecting instruments, lab systems, inventory, and data platforms, and building automated and agentic workflows that reduce manual activity for scientists.
• Landing the data: working with data engineering teams to get the right data in place and in the right shape, so solutions, analytics, and machine learning can be built on it.
• Getting to production safely and often: taking proven ideas through to secure, supportable, and compliant products, with infrastructure as code, CI/CD, automated testing, and monitoring built in so changes can be released quickly and with confidence.
• Building for reuse: leaving behind orchestration patterns, agents, and components that other teams across Labs & CMC can use.

Discovery, Solution Design, and Prototyping

• Work directly with Product Owners, scientists, and data engineering teams to understand lab workflows, the data involved, the constraints, and what a good outcome looks like.
• Pair with Product Owners to shape both the problem and the solution, challenge requirements where needed, and recommend the right technical approach.
• Design and build PoCs, PoVs, and MVPs in short cycles using real lab data and systems, with regular feedback from scientists.
• Look at feasibility, usability, data quality, and compliance early. Estimate the benefit in time saved, errors avoided, and manual effort removed, and recommend whether to continue, change direction, or stop.

Engineering and Delivery

• Write, test, and ship code, and own the technical design of the products you work on.
• Lead the move from MVP to production, covering architecture, integration, testing, security, deployment, monitoring, documentation, and support handover. Where the route is clear, ship it yourself.
• Keep technical debt under control and hand over cleanly to product and platform teams with working code, design notes, and known risks.

DevSecOps, Quality, and Observability

• Set up and maintain a streamlined SDLC for your products, with CI/CD pipelines that let the team release small changes often and safely.
• Build automated testing into every pipeline, including unit, integration, end-to-end, and performance tests, so quality checks run on every change and not just before a release.
• Build security in from the start through secure coding, dependency and container scanning, static and dynamic code analysis, and secrets management. Fix vulnerabilities promptly in line with GSK cyber security standards.
• Make sure solutions meet requirements for privacy, data integrity, and GxP and computer system validation where they apply. Where possible, use the pipeline to generate test evidence and audit trails automatically.
• Design observability in from day one, with logging, metrics, tracing, dashboards, and alerts using tools such as Grafana, Prometheus, and Elastic.
• Integrate products with ServiceNow for incident, change, and service management so they are ready for support and operational handover.
• Track delivery and service measures such as deployment frequency, lead time, change failure rate, and time to restore, and use them to keep improving how the team works.

Cloud Infrastructure

• Work with the GSK central Digital Platforms team to get the cloud infrastructure your products need, for example on Azure or GCP. Be clear about what is needed, review designs, and make sure it fits the solution.
• Use your knowledge of cloud networking, identity, access, and security to design solutions that run well on that infrastructure, and to work through issues when they come up.
• Where needed, provision and configure infrastructure yourself using infrastructure as code tools such as Terraform, following central platform standards and patterns.
• Package and deploy applications and services using container platforms and tools such as Kubernetes and Helm, with sensible settings for scaling, resilience, and cost.
• Manage development, test, and production environments through the pipeline rather than by hand, and keep them consistent with each other.

Labs & CMC Solutions

• Lab Workflow Orchestration: build and extend our lab orchestration platform, which links lab, data, and generative design systems, so scientists can run and monitor lab workflows from one interface.
• Lab Automation: design and build automation that connects instruments, lab execution, and data capture, moving towards closed-loop experiments where results feed straight into the next experiment design.
• Agentic Workflows: build AI agents and agentic workflows that take on manual, repetitive tasks for scientists and CMC teams, such as gathering data, checking results, and preparing documents.
• Landing the Data: work with data engineering teams to bring lab, analytical, CMC, and inventory data into well-structured data products, and build solutions on top of them that support analytics, traceability, and machine learning.
• Scientific Decision Support: use GenAI and machine learning to help scientists plan experiments, decide on next steps, and find relevant scientific knowledge.

AI-Native Engineering

• Use approved AI coding assistants and agent-based workflows for planning, coding, testing, review, and documentation as your normal way of working, while checking and correcting what they produce.
• Build and use the right harnesses for AI coding agents, such as reusable skills, project context and instructions, tool and MCP connections, and test and review loops, so agents produce reliable code with less rework. Share what works with the rest of the team.
• Design and build agent-based systems, retrieval-augmented generation, MCP integrations, and evaluation frameworks. Decide how much to automate and where scientists need to stay in control.
• Put human review steps, evaluation tests, guardrails, and monitoring in place so AI features stay accurate, explainable, and suitable for scientific and GxP use.

Technical Leadership, Governance, and Coaching

• Raise Responsible AI, data integrity, and validation questions at the prototype stage and work with Quality, Cyber, and Risk colleagues to resolve them.
• Add useful agents, orchestration templates, integration patterns, and reference designs to the Labs & CMC pattern library after each piece of work.
• Mentor and coach other engineers alongside your own delivery work.
• Use what you learn from working with the business to inform roadmaps, PI Planning, and portfolio priorities.

What You Bring

• At least 8 years of full-stack, software, data, or AI/ML engineering experience, including production systems where you were the technical lead or owner.
• A track record of designing and building working software quickly and then taking it into production. Experience in a startup, innovation team, or forward-deployed role is a plus.
• Strong Python skills and at least one of TypeScript/JavaScript, Go, Java, or similar, along with solid API, integration, and cloud engineering experience.
• Hands-on experience with GenAI, including agents, agentic and multi-agent workflows, retrieval-augmented generation, prompt engineering, and evaluation.
• Experience with workflow orchestration and automation tools (for example n8n, Airflow, or Temporal) and with working alongside data engineering teams on data pipelines and data products.
• Strong DevSecOps experience, including building CI/CD pipelines (for example GitHub Actions, Azure DevOps, or Jenkins), automated unit, integration, and performance testing, and security scanning built into the pipeline.
• Solid cloud infrastructure experience on at least one major cloud platform (for example Azure, GCP, or AWS), including networking, identity, and security, provisioning infrastructure using infrastructure as code (for example Terraform), deploying with container tools (for example Kubernetes or Helm), and working with central platform teams.
• Experience setting up observability with tools such as Grafana, Prometheus, and Elastic, and working with ServiceNow for incident and change management.
• Regular use of AI coding tools such as Claude Code or GitHub Copilot in your day-to-day development.
• Experience working directly with Product Owners and business users to shape requirements and drive a solution, and explaining options and risks clearly to senior stakeholders.
• Experience working with teams across different locations and time zones.
• Experience in a regulated environment, with a practical understanding of security, privacy, quality, and compliance.