Enterprise software, AI and automation.
Writing about how AI agents, automation and integrations actually land inside a serious business — no marketing, concrete examples, and the parts a proposal usually leaves out.
- Security10 min read
Security and privacy of AI agents: what to check before production
An AI agent in production has access to data and takes actions — which makes it a new attack surface. A checklist: permissions, prompt injection, tool guardrails and the fate of personal data.
Read - Architecture10 min read
Modernising a legacy system without stopping the business
A full rewrite fails because the old system keeps moving while you build the new one. Modernisation that works proceeds in small, reversible steps — one slice, behind a façade, while the business keeps running.
Read - AI & automation9 min read
How to measure the ROI of an AI project before you start it
The return on an AI project is not computed afterwards but before, against a baseline measured while nothing has changed. A guide to the costs people forget and a pilot that yields a defensible number.
Read - Fiscalization8 min read
E-fiscalization and ESIR: what the software must satisfy
Fiscalization is not a screen but the service behind it. Who the roles are in the model, what it means that an invoice must be fiscalized before it reaches the customer, and what happens when the network drops mid-sale.
Read - Integrations9 min read
ERP integration without double entry: what actually gets synced
Double entry is a symptom, not the disease. The disease is that two systems both believe they own the same record. The fix begins with deciding who owns what — and writing it down.
Read - Architecture9 min read
API-first: integration is an architecture decision, not a final task
Integration is rarely planned — it shows up as a request near the end of a project, once the most important decision has already been made in silence. API-first is the opposite order: the contract exists before the implementation.
Read - AI & automation9 min read
Voice AI agents: when they replace a call, and when they ruin it
A voice agent sounds human because of speed, not smarts. When it replaces a call, when it ruins one, and why containment must be measured honestly.
Read - AI & automation10 min read
RAG: AI that answers from your documents, not from memory
RAG is an approach where AI answers from your documents with a source, not from memory. How the pipeline works, why fine-tuning is usually not the fix, and when RAG is the wrong tool.
Read - AI & automation9 min read
Business process automation: from scripts to AI agents
Automation is not one technology but a ladder of five rungs. A guide to what each is good at, how it fails, and why you fix the process before you automate it.
Read - AI & automation9 min read
AI agents in the enterprise: where the ROI is, and where it isn't
The gap between a demo that impresses and an agent that survives production is not the model. It is which job you delegated, and who carries the cost when it is wrong.
Read