Coming soon
Where MCP belongs in a production architecture
MCP gives models a standard way to reach tools and data. The architectural question is where that boundary sits and who owns it.
- MCP
- AI systems
Insights
Practical notes on architecture and AI delivery.
No hype, no hot takes — working notes from designing and shipping real systems.
Articles
· 9 min read
RAG is a system, not a vector database
Why retrieval-augmented generation succeeds or fails on ingestion, evaluation, and operations — and what a production RAG architecture actually contains.
- RAG
- AI systems
- Architecture
Coming soon
Agentic workflow or ordinary automation?
A decision framework for when a workflow genuinely needs model judgement — and when a deterministic pipeline is the better system.
- Agents
- Automation
Coming soon
Architecture decisions a prototype tends to hide
Prototypes optimise for the demo. These are the decisions they quietly defer — and how to surface them before they become rewrites.
- Architecture
- Delivery
Curious what AI could actually do for your business?
Bring your questions, including the ones that feel too basic. In 30 minutes we go through how you work today, pick the task with the most to gain, and sketch what testing it would involve. No pitch, no obligation.
Book a free intro call
Prefer email? hello@nxtinno.com