Senior Full-Stack AI Engineer - Frontend & LLM-Enabled UX (Level 3) in London

GSK are seeking a Senior Full Stack AI Engineer with a strong focus on frontend engineering, user experience and product delivery. You will lead the design and implementation of immersive, interactive web experiences that leverage large language models (LLMs) as core product capabilities. In this role you will collaborate closely with product, UX/design, ML research and platform teams to translate user needs into elegant, reliable AI‐powered experiences and drive agile delivery from prototype to production.

Key Responsibilities

• Deliver highly interactive frontends (text‐based and video‐based chat experiences) built with modern frameworks like React that integrate LLM‐driven features (conversational interfaces, tool‐assisted workflows, structured editors, visualizations) to boost marketing productivity.
• Enable measurement / analytics to track usage of the app and of LLM features.
• Own the end‐to‐end product experience for LLM‐enabled features: collaborate with UX/design to craft flows, run rapid prototypes, and iterate based on user testing and analytics.
• Integrate LLMs into the application stack, handling prompt design, response parsing, tool use, chaining, and RAG strategies while balancing latency, cost and reliability.
• Implement frontend architectures and state management patterns that scale for complex, real‐time, and collaborative UIs (React hooks, Context, Redux/Zustand/Recoil, server‐state patterns).
• Build responsive, accessible, and performant UI components and design systems (component libraries, theming, design tokens) to ensure consistent product quality and fast iteration.
• Implement real‐time and progressive UX patterns (WebSockets, SSE, streaming LLM responses, optimistic updates, incremental rendering) that make AI interactions feel fast and human‐centered.
• Collaborate on frontend work with UX designers and researchers, follow standard coding conventions, and implement best practices for design, accessibility, testing and observability.
• Work with backend and platform engineers to define APIs, ingestion pipelines, vector retrieval, caching and rate‐limiting strategies that support robust LLM behavior.
• Drive agile delivery: break down features into deliverable increments, run sprint planning, pair with the tech PM on prioritization, and ensure production readiness through CI/CD, feature flags and canary releases.
• Define and instrument UX and model‐level metrics (engagement, latency, hallucination rates, user satisfaction) and use those signals to prioritize improvements.
• Ensure security and privacy are integrated into the UX (PII handling, consent flows, data minimization, redaction and access controls).

Minimum Requirements

• 6+ years building production web applications with proven ownership of user‐facing features; significant experience in frontend engineering and product delivery.
• Deep expertise with React (functional components, hooks) and modern frontend ecosystems including CI/CD.
• Demonstrable experience building highly interactive, immersive frontends (rich state, animations, canvas/visualizations, streaming UIs or collaborative editing).
• Strong UX sensibility and experience working directly with designers to translate wireframes and prototypes to polished, accessible product UIs.
• Hands‐on experience integrating LLMs / generative models into applications (OpenAI, Anthropic, local/open‐source LLMs): prompt engineering, response handling, safety guardrails, RAG architectures.
• Good understanding of how large language models work, including transformer architecture, tokenization, attention mechanisms, decoder/encoder‐decoder variants, fine‐tuning/few‐shot approaches, and common failure modes (hallucinations, bias, context‐window limitations).
• Solid backend experience (Node.js/TypeScript, Python, or similar) to collaborate on APIs, authentication, ingestion, and retrieval layers – enough to design pragmatic interfaces and own cross‐stack features.
• Practical experience building retrieval‐augmented generation pipelines (embeddings, vector stores, chunking/indexing).
• Proven track record delivering features in agile teams: sprint planning, incremental delivery, feature flagging, A/B tests and operationalizing metrics.
• Strong software engineering fundamentals: testing (unit/integration/E2E), observability, performance profiling, and security‐aware development.
• Excellent communication skills and experience mentoring engineers and collaborating across product, design and ML teams.

Preferred Requirements

• Familiarity with LLM orchestration and tools (LangChain, LlamaIndex, Haystack, OpenAI/Anthropic SDKs).
• Advanced visualization/graphics skills or experience building interactive applications.
• Experience with design systems and component libraries (Figma workflows, design tokens).
• Background in regulated industries (healthcare, pharma, finance) with awareness of compliance and data governance constraints.
• Contributions to open‐source projects, public talks or technical writing about LLM‐enabled UX or frontend engineering.