Senior Principal - Product Owner - Observability

Business Introduction

GSK remains committed to achieving bold commercial ambitions for the future. By 2031, we aim to deliver £40 billion in annual sales, leveraging our existing strong performance momentum to significantly increase our positive impact on the health of billions of patients globally. Our Ahead Together strategy is centred on early intervention to prevent and alter the course of disease, thereby protecting people and supporting healthcare systems.

Our diverse portfolio consists of vaccines, specialty medicines, and general medicines.

Through continuous innovation and a dedicated focus on scientific and technical excellence, we strive to develop and launch new, groundbreaking treatments that address critical health challenges.

Job Purpose

The Senior Principle Owner – Observability owns GSK's enterprise observability platform (observability.gsk.com) as a product. This is a techno-functional role. It sets the product vision and owns and orders the backlog, and it is hands-on enough to configure BindPlane and OpenTelemetry pipelines, write parsing rules, script ingestion utilities, build Grafana dashboards and debugging issues.

The role leads the platform's back-end migration from self-managed Prometheus/Mimir and Elastic SaaS to Google Cloud Observability. Telemetry collection moves to BindPlane (Google Edition), and Grafana remains the single interface for GSK teams. Beyond the migration, the role turns into managing backlog, licensing & transforming platform with AI capabilities into services that business teams use directly: plain-language questions about service health, early warning on business-critical processes, and clear incident and health summaries.

Success is a platform that teams adopt by choice: onboarding through golden templates rather than tickets, SLOs that reflect what the business needs, and alerting that cuts noise rather than adding to it. The role treats consuming teams as customers, because they can route around a platform that is hard to use. It links platform work to D&T and business outcomes, and makes the cost of deferred technical debt, upgrade and instrumentation work visible.

Key Responsibilities

1. Platform ownership – Run observability.gsk.com as a product with a clear vision, roadmap, SLOs, capacity plan, run-cost model and lifecycle (patching, upgrades, end-of-life). Own and order the backlog, linking each item to platform OKRs and stating the cost of not doing it.

2. Google migration – Deliver the back-end migration from Prometheus/Mimir and Elastic SaaS to Google Cloud Observability (Cloud Logging, Managed Service for Prometheus, Cloud Trace) before the Elastic contract ends in April 2027. Sequence waves by business criticality, run old and new in parallel, and prove parity for core metrics, dashboards, alerts and log searches. Keep the Grafana experience unchanged, and decommission legacy components without loss of coverage or audit history.

3. Reliability – Achieve 99.9% availability for tier-1 platforms. Agree SLOs with service owners, close coverage gaps so reliability is measured rather than inferred, and lead incident response, blameless post-incident reviews and problem management. Replace repeated manual recovery with permanent automated fixes.

4. Telemetry pipelines and data quality – Own how telemetry reaches the platform. Run the OpenTelemetry agent fleet through BindPlane (Google Edition), covering sources, processors, destinations and staged config rollout. Own parsing and normalisation (Grok and regex, structured JSON logging, consistent labels and resource attributes), the Python and OTel utilities that ingest end-user services (EUS) telemetry, synthetic monitoring, and Google Data Fusion feeds for network telemetry. Control ingest cost through sampling, filtering, exclusions, retention and cardinality management.

5. Architecture and scaling – Partner with the Observability/Platform Architect on how the platform scales: BindPlane gateway and collector topology, Google Cloud project, metrics-scope and log-bucket design, log routing, multi-region and hybrid (GCP, Azure, on-premise) coverage, cardinality and retention. Translate business demand into capacity and architecture decisions, and explain the trade-offs to business stakeholders in plain terms.

6. AI-powered observability for business teams – Offer AI in observability as a product that R&D, Manufacturing, Supply and Commercial teams use directly: plain-language questions about service health, proactive early warning, and AI-generated incident and service-health summaries. Build the unified data model across metrics, logs and traces that underpins this and the AI for Tech platform, using BigQuery, BigQuery ML, Vertex AI and Gemini, with governed use of third-party LLMs. Route alerts to ServiceNow ITOM and the AIOps platform, and maintain Grafana plugins for external sources such as ServiceNow and Oracle.

7. Everything as code, secure by default – Manage collector configs, parsers, log sinks, alert policies, dashboards and access in Git through IaC (Terraform Google and Grafana providers, BindPlane config, Ansible) and GitOps; no configuration lives only in a console. Build in SSO, least privilege, project isolation and masking of sensitive data from the start.

8. Observability templates and developer experience – Build reusable OTel instrumentation templates, BindPlane configs, parsers and Grafana dashboard packs so delivery teams onboard and self-serve without tickets. Measure and improve DORA metrics and platform NPS, and provide onboarding, documentation, migration guides and office hours.

9. Leadership – Set standards, review designs and code, and mentor engineers, including partner engineers. Manage vendor and partner relationships (Google, observIQ/BindPlane, Grafana Labs, Elastic through exit, ServiceNow, Red Hat, GitHub, Microsoft), holding partner-delivered work to internal standards. Apply the Platform Architect's reference architectures and raise a case for change rather than diverging locally.

Experience Required:

Experience

• Overall, 15 Years and 12+ years in platform, infrastructure or software engineering, with hands-on platform experience

• 5+ years owning a production platform end-to-end at enterprise scale (5,000+ users), including product ownership of its backlog and roadmap

Observability

• Hands-on Google Cloud Observability in production: Cloud Logging (log router, sinks, buckets, log views, Log Analytics, Logging Query Language), Cloud Monitoring and Managed Service for Prometheus (PromQL, alerting policies, metrics scopes), and Cloud Trace

• Hands-on Prometheus/Mimir and Elastic (Elasticsearch, Logstash, Kibana, ingest pipelines), enough to run the legacy stack and exit it safely

• Grafana in depth: dashboards as code, alerting, data sources, plugins, folders and RBAC

• OpenTelemetry: SDKs, collector, semantic conventions and resource attributes

• Defines SLOs and alerting that cut noise, with evidence of before-and-after impact

Telemetry pipelines and parsing

• BindPlane (Google Edition preferred) or equivalent OTel fleet management: agents, sources, processors, destinations and staged config rollout across large fleets

• Writes and debugs Grok patterns and regex that turn unstructured logs into structured fields (Logstash, Elastic ingest pipelines, BindPlane/OTel processors), and moves teams to structured JSON logging

• Builds Python ingestion utilities for non-standard sources, such as end-user services, using OpenTelemetry

• Applies sampling, filtering, enrichment, PII masking and multi-destination routing, including telemetry shared with a SIEM

• Controls ingest cost through log exclusions, retention, metric cardinality and sample volume

Scripting and programming

• Strong Python (ingestion utilities, OTel SDK, automation) plus Go or PowerShell; comfortable with Bash

• REST APIs and webhooks (Google Cloud, Grafana, BindPlane, ServiceNow, Elastic) for automation and inte

View and apply on HireChats