/ Service
Without an evaluation set, retrieval tuning is opinion and every change is a gamble. I build the measurement first — what counts as a correct, grounded answer — so changes to chunking, search and reranking become comparable engineering decisions.
/ What this covers
/ Evidence
Retrieval architecture · ReAct agents
Architected and delivered a scalable enterprise RAG and agent platform: a 500GB+ embedding and ingestion pipeline feeding hybrid search with reranking and vector retrieval, behind a ReAct agent architecture. Retrieval quality was treated as the product — measured chatbot accuracy improved to 95% and response latency was reduced.
Grounded assistant · Procurement
Designed and productionised a grounded enterprise Procurement assistant on Azure AI Foundry, using LangGraph for orchestration, ChatKit for the interface, PGVector for retrieval and integrations with MCP servers. Answers stay grounded in trusted enterprise documents, and the architecture was standardised into a reusable reference implementation.
Read the Enterprise Procurement Knowledge Assistant case study
/ Related services
ReAct agent architectures and grounded assistants, with tool access, guardrails and evaluation designed in.
Paved paths and internal platforms so AI applications start from a supportable, governed baseline.
Governed agent access to internal systems through MCP servers and shared integration patterns.
The operating model — identity, approvals, usage and cost controls — that lets AI scale past pilots.
/ Next step
Open to full-time Senior–Staff AI / agent platform roles (remote-friendly), as well as contract and consulting engagements. Response within two business days.