Designing
Intelligent
Systems.
I design and document agentic AI, multi-agent architectures and the infrastructure that takes them into production.
Hermes
A low-latency orchestration platform that transforms live speech into reliable AI conversations through streaming speech recognition, agentic reasoning, retrieval, and neural speech synthesis.
Athena
A production-oriented retrieval platform designed for trustworthy enterprise knowledge access through optimized search, reranking, vector search and permission-aware retrieval.
Aegis
An enterprise AI governance platform focused on explainable reasoning, calibrated confidence estimation and policy-aware decision validation.
Multi-Agent Systems
Coordinating specialised agents through orchestrators, planners and shared state. Patterns for delegation, conflict resolution and handoffs.
Agent Memory Systems
Short-term working memory, episodic recall, vector-based semantic memory and long-horizon context compression strategies.
RAG Architectures
From naive retrieval to hybrid, agentic and graph RAG. Choosing the right retrieval substrate for your domain.
MCP Architectures
Model Context Protocol patterns for tool servers, resource exposure and cross-process agent capability sharing.
Harness Engineering: The Missing Reliability Layer in Agentic AI Systems
Harness Engineering introduces a runtime discipline for building reliable, production-grade agentic AI systems through deterministic execution boundaries.
The Shape of Agentic Systems
Most agent failures are not model failures — they are shape failures. A taxonomy of the loops, hierarchies and graphs that actually work in production.
Memory Is the System
Agent capability is mostly a function of memory design. A breakdown of working, episodic, semantic and procedural memory in practice.
RAG Is a Retrieval Problem, Not a Generation Problem
Why most RAG systems plateau, and the retrieval-side investments that actually move the quality curve.
Evals as Product Engineering
Treating evaluation as a first-class system — datasets, judges, traces, regressions and the discipline of continuous quality.
Memory-first agent design
Treating episodic and semantic memory as the core architecture, not a bolt-on.
Evaluation as infrastructure
Continuous eval harnesses that run with every change, not just at launch.
Context engineering
Designing the information environment an agent operates inside — instructions, retrieval, tools, state.
Let's build something
worth documenting.
Whether it's a system architecture review, a writing collaboration, or an idea you want to stress-test — I'm reachable.