Inferex: An LLM inference engineering testbed
A hands-on LLM inference stack combining a FastAPI gateway, quantisation benchmarks, vLLM and SGLang engine comparisons, observability, and Kubernetes deployment on a single GPU workstation.
Writing
I write to connect research and engineering: sharing what works, the decisions behind it, and what each world can learn from the other.
Platframe: Building an AI-Assisted Investigation Platform
A walkthrough of the environment, architecture, workflow, subsystem boundaries in Platframe. A multi-stage AI workflow system using insurance investigations as the operational environment.
Evaluation as a System Primitive for Model Behaviour
Embedding evaluation into execution pipelines to enable consistent comparison of model behaviour across API and local providers.
Agent Learning: Execution-Aware Policy Optimization for LLM Tool Systems
Tool orchestration in LLM systems as a learnable policy optimized for execution cost, latency, and constraint satisfaction.