Senior Forward Deployed Engineer (LLM, RAG & Agentic AI)

Job ID: 11756Location: London, England, GBType: PermanentSalary: USD excellent salary, bonus, full bens per yearPosted: 2026-10-29Senior Forward Deployed EngineerLLM, RAG, Agentic AI and Prompt Engineering | Permanent | Remote first, UKA fast-growing, engineering-led AI consultancy is hiring Lead and Senior Forward Deployed Engineers on a remote-first basis across the UK. You will lead the design and delivery of complex client engagements spanning agentic systems, retrieval architectures and semantic layers, working directly inside enterprise client teams.Client demand for AI-led transformation is accelerating and our client is expanding its permanent engineering team across agentic systems, retrieval, ontologies and AI-enabled operations. As a Forward Deployed Engineer, you will work closely with the client, in their workshops and inside their systems, turning real operational problems into working AI solutions.These are senior, hands-on consulting roles that combine deep engineering craft with commercial and product thinking. You will work with senior client stakeholders alongside consulting and delivery colleagues, and lead technical design on your engagements. Prompt design, evaluation and guardrails are central to how you build, so the agents you ship are safe, reliable and measurable once they are live.You will shape both client outcomes and the firm’s long-term AI engineering capability, while staying hands-on in the code.What you’ll doEmbedded client delivery and technical leadershipLead technical design within client engagements and set architectural direction across delivery teamsWork alongside client teams, remotely and on occasional site visits, to understand their workflows, data and constraints first handTurn ambiguous client problems into clear, feasible engineering approaches and fast prototypes for client validationAssess client data, platforms, security constraints and operating modelsContribute to framing, use-case shaping and technical scoping discussionsWork with client domain experts to surface edge cases and operational realitiesProduce clear, lightweight technical artefacts for client and internal audiencesAct as the trusted technical voice on the engagement, from first workshop to go-liveAgentic AI, LLM and RAG engineeringDesign and build LLM applications using modern orchestration patterns, prompt frameworks and evaluation loopsBuild multi-agent architectures covering planning, delegation, tool use, memory, safety constraints and monitoringBuild retrieval and vector-based systems: embeddings, RAG, structured reasoning and semantic workflowsConnect agents to enterprise platforms through APIs, tools, MCP servers and connectorsApply ontologies or knowledge graphs where they improve reasoning and data alignmentPrompt engineering and evaluationDesign, test and version system prompts, tool descriptions and agent instructionsBuild reusable prompt libraries and patterns that hold up across clients and use casesCreate evaluation sets and automated eval loops to measure accuracy, safety and consistencyImplement guardrails, structured outputs and explainability checksDiagnose and fix failure modes such as hallucination, tool misuse and prompt injectionDelivery excellence, AI operations and reliabilityImplement evaluation frameworks and safety checks across models and agentsDesign monitoring, logging, tracing and incident-response patterns for live agentsApply governance, risk and compliance principles within client environmentsSupport releases, environments and handover into client operationsEnsure reliability, reproducibility, performance and cost controlBuilding team capabilityTurn field learnings into reusable agent patterns, templates and acceleratorsFeed client insight back into the wider engineering team and its standardsCoach peers and share new research, tools and techniquesWhat you’ll bringEssential6 to 8 years in software engineering, AI engineering or applied data engineeringStrong hands-on experience with LLMs, embeddings, RAG, retrieval stacks and vector databasesExperience designing or implementing multi-agent systems or tool-calling frameworksStrong prompt engineering skills, including system prompt design, tool use and structured outputsExperience building evaluation frameworks for LLM or agent behaviourStrong Python, with experience building production-grade systemsAt least one major cloud AI ecosystem: Azure OpenAI, Google Vertex AI, AWS Bedrock or AnthropicFamiliarity with semantic modelling, ontologies or knowledge graph conceptsProven ability to rapidly prototype solutions for client validationExperience working directly with clients in consulting or professional servicesDesirablePrevious forward deployed, solutions engineering or technical leadership experienceCommercial awareness and product thinking, including scoping and estimating technical workAgent frameworks such as LangGraph, CrewAI, AutoGen, Semantic Kernel or the OpenAI and Anthropic agent SDKsLLM evaluation and observability tools such as Langfuse, LangSmith, Arize or BraintrustDeeper ontology or knowledge graph expertiseDomain experience in life sciences, pharma or healthcare, FMCG or retail, insurance, or banking and financial servicesStrong engineers from non-AI backgrounds who have moved quickly into agentic AI and can show what they have built are very welcome.