I design and deliver enterprise-grade AI and IT systems — from local-first LLM infrastructure and edge inference to cloud architecture and strategic technology consulting.
End-to-end AI system design — from on-device NPU inference and RAG pipelines to production-scale LLM deployments. Optimised for latency, cost, and data privacy.
Resilient infrastructure for modern organisations. ARM64 edge computing, hybrid cloud architecture, and security-first design at every layer.
Technology-driven organisational change. System design, process automation, and strategic advisory that connects IT investment to measurable business outcomes.
Designed and deployed a complete local AI inference stack on ARM64 hardware. NPU-accelerated LLM serving via GenieAPIService, Ollama multi-model management, and LiteLLM as a unified OpenAI-compatible proxy — all systemd-managed with automated recovery.
Full-stack intelligent agent platform with multi-model orchestration, autonomous workflow execution, and Supabase-backed persistence. Multi-provider AI routing with context management and real-time streaming.
Multi-node Proxmox cluster with Ceph distributed storage, automated boot-recovery procedures, Immich ML with remote inference offloading, and a hardened security posture across all nodes.
Whether you're scoping an AI architecture project, need strategic IT advisory, or want to discuss emerging technology — I'm open to select consulting engagements and collaborations.