Bashar Ayyash

AI Engineer & Full-Stack Tech Lead Helping Businesses Grow with AI

I'm a Tech Lead and AI Engineer based in Amman, Jordan, specializing in Laravel, Next.js, React Native, and AI agents (RAG systems) for MENA businesses. Bridging the gap between 20 years of expertise and modern AI Agents.

CV

AI Architecture

Building intelligent RAG systems and autonomous agents that actually solve business problems.

const agent = new Agent("Yabasha");
await agent.optimize("Your_Workflow");
Base

Amman, Jordan 🇯🇴

Available for Consulting

The Arsenal

LaravelNext.js 16BunReact NativeTailwind v4OpenAI APIDockerPostgreSQLTypeScript
20+Years Experience
50+Projects Shipped
MENARegion Focus
AI + CodeFull-Stack Delivery

Who I Help

I work with CTOs, product leaders, and founders in the MENA region who need a senior engineer to design and ship AI-powered products: RAG assistants, Laravel + Next.js platforms, and React Native mobile apps.

  • •Teams shipping React Native (Android/iOS) fintech apps: payment gateway integrations, secure flows, and real-world performance on older devices (startup time, memory, battery) with latency reduction where it matters.
  • •NGOs / humanitarian programs running market monitoring in high-risk areas: web + tablet + mobile collection, offline-first sync, cleaning millions of records, and donor-ready reporting from reliable pipelines.
  • •Operations-heavy businesses needing multi-database / clustered backends: data partitioning, queues, secure edge nodes, and resilience when the network doesn’t.
  • •Restaurants and chains modernizing bookings: mobile/tablet/cashier workflows, handling booking calls, and analytics dashboards that turn demand into staffing and revenue decisions.
  • •HR and procurement workflows: resume/profile management, structured evaluations, and tender submissions with search, permissions, and auditability.
  • •Supply chain & distribution platforms: factory → sub-distributor flows for booking, acceptance, storage, billing, and reconciliation across roles and locations.
  • •Product teams needing small-but-critical utilities that scale: internal URL shorteners, automation scripts, and “boring” systems that must be fast, reliable, and secure.
  • •Builders and creators shipping to the public: wasansart.com, plus developer tooling such as GEX CLI, copywrite-skill, and modernizing a Laravel AI SDK (including tool-using agents).
Read the Guides

Web & Mobile Platforms

Robust, scalable platforms using Laravel for backend and Next.js for high-performance frontends.

Business Automation

Native-feel iOS and Android applications built with React Native and Expo.

AI Chatbots & Assistants

Production RAG assistants and tool-using agents that answer from your docs, tickets, and databases — with citations, access control, and evaluation. Read the RAG guide.

How We Work

Clear scope, fast feedback loops, and production-quality delivery—without ceremony.

  1. 1. Scope & Success Metrics (2–5 days) — clarify the journey, constraints, risks, and define KPIs/SLAs/acceptance criteria.
  2. 2. Architecture & Plan (3–7 days) — choose the approach, map milestones, and lock the first deliverable.
  3. 3. Build the Vertical Slice (1–2 weeks) — deliver an end-to-end slice (UI → API → DB → observability) to validate direction early.
  4. 4. Iterate to Full Feature Set (2–6+ weeks) — ship weekly increments: features, edge cases, performance work, and stakeholder feedback loops.
  5. 5. Hardening, Launch & Ongoing Improvement (1–2 weeks + ongoing) — testing, security review, monitoring, rollout plan, and handover docs.

Production-Grade Standards

I optimize for reliability and maintainability—so the system keeps working when it's under load, audited, or handed to a team.

  • •Testing by default: unit + integration + e2e where it pays off; realistic fixtures; regression coverage for critical flows.
  • •CI/CD discipline: automated lint/type-check/tests, preview builds, safe migrations, and repeatable deployments.
  • •Observability built-in: structured logs, metrics, traces, alerts, and dashboards tied to business KPIs (not just CPU).
  • •Security-first engineering: least privilege access, secrets management, input validation, audit logs, and threat-aware reviews.
  • •Performance budgets: latency targets, caching strategy, background jobs/queues, and profiling—not "optimize later."
  • •React Native quality bar: crash-free sessions, offline handling, slow-device testing, startup performance, and app-store-ready release processes.
  • •Data reliability: idempotent sync, conflict handling, backfills, and safe reprocessing for large-scale datasets.
  • •AI safety & evals: prompt/version control, eval datasets, hallucination/grounding checks, tool-use guardrails, and human approval gates where needed.

FAQ

Use RAG when your answer must be grounded in changing or proprietary knowledge (docs, tickets, policies) and you need citations. Consider fine-tuning when you want consistent style/format, domain-specific behavior, or tool-selection patterns—and your knowledge is stable. Many real systems use both: RAG for facts, light tuning for behavior.

Typically by adding an AI service layer (queues, retries, rate limits), a small adapter for providers, and an audited tool/API boundary. We start with 1–2 high-value workflows, add observability and evals, then expand. The goal is incremental adoption: minimal disruption, measurable outcomes.