Engineering moves fast. The frameworks, patterns, and tools that were best practice eighteen months ago are sometimes obsolete today — and occasionally what was considered over-engineering is now table stakes.
Icon Software builds software for organizations that depend on it: ERPs for businesses where the system is the operation, security infrastructure for networks that can't afford a breach, custom applications for industries where off-the-shelf doesn't fit. The decisions we make on those projects are grounded in practice, not theory.
This blog is where we share what we're learning.
What this journal covers
We write about eight topics, each one reflecting a domain where our work intersects with current technology change:
Agentic AI & Multi-Agent Systems. How to architect, operate, and govern AI agents in production. Not prompting experiments — production-grade systems with observability, guardrails, and rollback paths.
AI-Native Software Development. How development workflows change when AI is a first-class participant: code generation, automated review, test synthesis, and the engineering discipline to use these tools without accumulating invisible debt.
Cloud 3.0 & Hybrid Multi-Cloud. Infrastructure architecture for organizations operating across cloud providers and on-premises environments. Cost, compliance, and operational complexity are all in scope.
Cybersecurity & Post-Quantum. Practical security engineering — zero-trust architectures, identity-centric access controls, and preparing for the cryptographic transition that quantum computing will eventually force.
Low-Code / No-Code Platforms. Where these platforms create genuine leverage for business teams, where they create technical debt, and how to integrate them with the broader application estate.
CI/CD & DevOps Automation. Pipeline design, deployment strategies, environment parity, and the operational patterns that make frequent releases reliable rather than stressful.
Physical AI & Robotics. Edge computing, embedded systems, and AI at the physical layer — relevant to logistics, manufacturing, and field service operations.
AI Sovereignty & Tech Governance. Policy, regulation, and organizational governance for AI systems. Especially relevant for organizations in regulated industries and public sector.
What you won't find here
Generic tutorials that restate documentation. Posts that hype technology without addressing its failure modes. Vendor marketing disguised as engineering opinion.
The standard we hold ourselves to: every post should give a practicing engineer something they can use or a decision-maker something they can act on. If a topic doesn't meet that bar, we don't publish it.
How we use this in practice
The posts here reflect real decisions on real projects. When we write about zero-trust architecture, it's because we've implemented it in environments where the consequences of getting it wrong are significant. When we write about ERP implementation pitfalls, it's because we've watched organizations repeat the same avoidable mistakes at considerable cost.
We're a team based in Amman, building software for organizations across the region and beyond. The technical problems are universal; the context sometimes isn't. We'll note when regional specifics — regulatory, infrastructure, or organizational — shape the advice.
Subscribe
Posts appear in RSS as they're published. No newsletter, no tracking, no noise.
Start with CI/CD pipeline patterns that survive real teams if you're looking for something immediately practical, or browse the insights hub by topic.