Currently architecting GenAI for a global aerospace manufacturer
Enterprise GenAI, engineered for production. I architect RAG, multimodal and agentic systems for large enterprises — and own them from proof of concept and proposal through to a governed, observable release.
<90 s to answer across 1M+ enterprise emails — down from about 30 minutes <2 min to generate a project charter that took 30–40 effort hours 300+ users on a production GenAI platform I architected 8 AI and data solutions grown from one proof of concept Delivered in
Manufacturing Aerospace, Steel Energy & Utilities Electricity, Gas Travel Corporate, Leisure For clients across the UK, Canada and Australia.
Selected work
Production systems that replaced the way people actually work. Featured · Global aerospace engine manufacturer
GenAI email search & summarisation Problem Engineers lost around 30 minutes per query digging through a million historical emails with slow, inaccurate mailbox search.
Approach Hybrid retrieval — semantic vectors with BM25 keyword matching — behind LLM guardrails, with AI summaries shown next to the source email's metadata.
Outcome Under 90 seconds per query, released to production with a cleared quality gate and full observability.
My role AI Solution Architect
Delivery Team of 5
Status In production Deployment view Read the case study → Capabilities
Architecture that survives security review, cost scrutiny and real users. Agentic & GenAI architecture LangGraph and LangChain workflows, Azure OpenAI, multimodal and vision LLMs, prompt and context design.
Retrieval engineering Hybrid vector + BM25 search, chunking and ranking, latency work on corpora of a million documents and up.
Enterprise hardening Guardrails, access control, SonarQube quality gates, observability and MLflow‑based LLMOps on AKS.
Data science foundations Causal inference with DoWhy, change‑point detection, statistical process control, classical ML.
