Leads the team this role joins
Decides on early hires
Runs hiring at the company
From your own Gmail and LinkedIn.
About the role
This is a senior individual contributor role for an architect focused on AI solution design, indexed as based in Bengaluru, within GSK's AI, ML and Cloud Security function. The architect designs the solutions that AI engineering builds, covering LLM orchestration, agent design, retrieval-augmented generation and security for AI services. The function sits within the Chief Security Office and spans the UK, India and the US, and the role holds design authority without line management.
What you’ll do
- Design AI solution architectures covering LLM orchestration, agent patterns, RAG and retrieval boundaries, embedding security and API security.
- Validate the agentic framework design before handoff and confirm it is buildable, testable and operable.
- Write pattern library entries that engineering teams build from.
- Lead architecture reviews of new AI tool proposals against standards.
- Contribute to data fabric architecture and the controls at each classification boundary.
- Produce specifications detailed enough for engineering to build without reinterpretation.
- Apply AI security standards and threat models, and escalate designs that cannot meet them.
- Assess the feasibility of business and R&D use cases.
- Own the AI architecture target state and roadmap for the domain.
- Review agent and RAG designs from other architects and sign them off or return them.
- Apply GxP and data integrity requirements to AI designs.
What they’re looking for
- 15+ years in security or solution architecture, with a recognised architecture certification
- Alternatively, 8-12 years in security or solution architecture with an AI architecture focus at enterprise scale
- Experience designing agentic frameworks, including trust boundaries, tool-calling scopes and human-in-the-loop controls
- AI threat modelling and secure AI data flow design, covering vector store access and AI-specific threats in the OWASP LLM Top 10 and MITRE ATLAS
- Experience assessing AI tools against security standards, including model provenance, data residency, classification and third-party risk
- Ability to write architecture specifications that engineering can build without rework
- Ability to whiteboard an agentic or RAG architecture you designed, including prompt injection and data classification handling
- A bachelor's degree in a relevant discipline, or equivalent professional experience
Nice to have
- Pattern library and architecture governance experience
- EU AI Act alignment and AI regulation in regulated industries
- Azure OpenAI and Vertex AI private endpoint and customer-managed key patterns
- Experience with designs handed to a third party to operate
- Cloud AI services architecture across Azure OpenAI, Vertex AI and private model hosting
- SABSA or TOGAF certification
Skills
Questions about this role
Where is this role based?
The listing is indexed as Bengaluru, and the wider team spans the UK, India and the US.
Does the role manage people?
No, it is an individual contributor role with no line management.
How much experience is required?
The listing asks for 15+ years in security or solution architecture, or 8-12 years with an AI architecture focus.