AI Software
AI-powered applications, intelligent interfaces, AI-assisted workflows and software systems.
Luther Sabuero
Vibe Coding • AI Agents • Automation • Full-Stack Systems
I turn ideas into production-ready software using AI-native engineering workflows.
From natural-language ideas to architecture, code, testing and deployment — I use AI as an engineering partner to build intelligent software systems.
Think. Architect. Vibe Code. Engineer. Deploy.
What I Build
Six focus areas where AI-native engineering meets real-world systems.
AI-powered applications, intelligent interfaces, AI-assisted workflows and software systems.
AI-native development using structured prompts, iterative coding, architecture, testing and refinement.
Intelligent agents capable of reasoning through workflows, retrieving information and performing structured tasks.
Business and technical workflows transformed from repetitive manual processes into automated systems.
Cloud platforms, VPS, networking, deployment architecture, systems and infrastructure.
Modern web applications, APIs, databases, dashboards, business platforms and integrations.
About
With more than 20 years of experience in IT engineering, I have worked across infrastructure, networking, cloud platforms, enterprise systems, web technologies, automation and business operations.
Today, I combine that engineering foundation with AI-native development and Vibe Coding to build software faster while maintaining structure, reliability, security and maintainability.
My focus is not simply writing code. It is understanding a problem, designing the right system and using AI to accelerate the entire engineering lifecycle.
Career Evolution
IT Experience
The engineering foundation behind the AI-native work — Microsoft Azure, Microsoft 365, Google IT, networking, enterprise systems, security and infrastructure.
Career Timeline
Building and maintaining physical servers, virtual machines, an in-house data center, workstations and base IT operations.
Designing and operating networks, enterprise systems, communications and multi-site IT environments.
Migrating and operating workloads on Microsoft Azure, cloud VMs, Microsoft 365 and Office 365.
Building web applications, automating business and technical workflows, and moving into software engineering.
Combining 20+ years of IT engineering with AI, LLMs, agents and Vibe Coding to build production software.
Employment History
IT Support
Foundational IT support role — desktop, hardware, basic systems and end-user troubleshooting.
IT
IT operations across point-of-sale and business technology environments.
IT
IT operations in a semi-government sector organization — systems, networking and user support.
IT & Software Engineering
Current role — IT infrastructure, enterprise systems, cloud and AI-native software engineering for URUK Group, including the live EPC Opportunity Intelligence platform.
Platforms & Vendors
Credentials & Training
IT support, systems administration, networking, security and troubleshooting foundations.
Full CV available on request.
Featured Projects
Real systems built with AI-native workflows — from trading intelligence to off-grid infrastructure. Click any project for the full case study.
Deployed Work
Real, shipped projects you can visit. Each link opens the live deployment — these are production systems, not screenshots.
Methodology
Vibe Coding is not random prompting and accepting whatever AI generates. My approach combines natural-language interaction with software architecture, engineering standards, testing, security and continuous review.
Define the problem.
Use AI to investigate possible solutions.
Define system boundaries and architecture.
Use AI coding agents to implement.
Validate functionality and assumptions.
Challenge AI-generated implementation.
Improve quality and maintainability.
Move the system into production.
Observe and improve continuously.
AI makes software development faster. Engineering makes it reliable. Vibe Coding connects the two.
AI Agent Workflow
AI can act as several engineering roles through the lifecycle. The human remains responsible for final engineering decisions.
Explores architecture and technical decisions.
Implements features and components.
Creates tests and identifies potential failures.
Challenges implementation and detects issues.
Maintains technical documentation.
Supports deployment and infrastructure workflows.
AI accelerates every role. The human owns architecture, security and production decisions.
How I Build
Ten principles that keep AI-assisted development reliable, secure and production-ready.
Architecture before code.
Solve the problem, not just the prompt.
AI accelerates engineering; it does not replace engineering judgment.
Keep systems modular.
Automate repetitive work.
Test continuously.
Security by design.
Document important decisions.
Build for maintainability.
Production reliability over prototype hype.
AI Engineering Lab
A living view of what is being built, researched and explored. Status labels reflect real state — no inflated claims.
ACTIVE DEVELOPMENT
Multi-timeframe XAUUSD analysis engine with AI reasoning and a 12-step decision pipeline.
ACTIVE
The AI-native engineering methodology applied across every project.
DEPLOYED / ACTIVE
Live platform monitoring 61+ Iraq energy sources for EPC opportunities at leads.urukepc.com.
EXPLORATION
Turning repetitive business workflows into automated, monitored systems.
PROJECT
Headless commerce architecture built with AI-assisted workflows.
DEPLOYED / PROJECT
Solar-powered off-grid PtMP wireless network for remote field equipment.
Technology Stack
Organized by role rather than a logo wall. Only technologies that can reasonably be supported by my actual experience.
AI
Architecture Gallery
Click any architecture to open a detailed explanation of how the system is structured.
Business + Engineering
My work sits between a business problem and a business outcome. Engineering is the bridge — AI and automation are how the bridge is built faster.
Understanding business requirements is not optional. The most elegant code that solves the wrong problem is still a failure. This is why engineering judgment matters more as AI accelerates implementation.
Problem → Outcome
Contact
Have an idea, workflow or business problem that could become software? Let's turn it into an engineered system.