Skip to content
ZAF

Company

ZAF builds the technology, integrations, data infrastructure and AI systems businesses need to operate and grow.

We are a technology, AI and growth solutions provider serving businesses in Malaysia. Not a web design agency, not an ad agency — an engineering partner that connects software, data, AI and marketing to business outcomes.

Positioning

Technology + Integration + Data + AI + Growth.

Most providers cover one of these. We cover all five, because the value of each depends on the others: software without integration creates silos, data without integration is incomplete, and AI without good data is guesswork.

What we build

  • Business systems and web applications

    CRM, operations platforms, portals, SaaS products and internal tools designed around your process.

  • Integrations and middleware

    Secure connections between CRM, ERP, payments, WhatsApp, Meta, Google and internal systems.

  • Data infrastructure and dashboards

    Pipelines, reporting models and executive views that join marketing, sales and finance.

  • AI agents and automation

    Assistants, document processing, lead intelligence and workflow automation with human approval.

  • Marketing technology

    Conversion tracking, Conversions API, attribution and AI campaign analysis linked to revenue.

Our approach

Discovery before design. Architecture before code. Measurement from day one.

  1. 01

    Discover

    Understand the business, the problem and the objectives before proposing technology.

  2. 02

    Architect

    Design the system, integration points and data flow so everything connects from day one.

  3. 03

    Build

    Develop the solution in short cycles using the appropriate, well-supported technologies.

  4. 04

    Integrate

    Connect platforms, APIs and business data with authentication, validation and logging.

  5. 05

    Launch

    Test, secure and deploy with monitoring, documentation and training in place.

  6. 06

    Improve

    Measure results, review the data and keep improving the system after launch.

Engineering philosophy

Boring technology, careful architecture, honest scope.

We choose well-supported, widely understood technologies over novel ones, so systems remain maintainable by any competent team — including one that is not us.

We design the data model, access control and integration points before the interface, because those are the decisions that are expensive to change later.

We prefer a small first release that works over a large one that is late. Every phase has a defined outcome, and we say so when a custom build is not the right answer.

  1. 01Business-First EngineeringTechnology decisions begin with the business problem and the outcome it needs to produce.
  2. 02Systems, Not SilosWe connect software, data and platforms so information moves without manual hand-offs.
  3. 03Built for MeasurementAnalytics, tracking and attribution are designed in from the start, not added after launch.
  4. 04AI With PurposeAI is implemented where it improves real workflows and decisions, with human approval where it matters.
  5. 05End-to-End CapabilityFrom interface and backend to integrations, analytics and automation — one accountable team.

Why integration matters

Systems should work together, not operate in silos.

A CRM that does not know which ad produced a lead, an accounting system that does not know which campaign produced an invoice, a WhatsApp inbox disconnected from both — this is the normal state of most businesses, and it is why reporting is never trusted.
  • Data entered once and used everywhere — no double entry, fewer errors
  • Attribution preserved from first click to invoiced revenue
  • Automation triggered by real events in real systems
  • AI that reasons over complete data rather than one platform's view
How we approach integration

How we approach AI

AI with purpose, grounded in your data, approved by people.

We apply AI where it improves a real workflow or decision — reading, classifying, summarising, analysing — and keep deterministic rules in code. Recommendations are explained, logged and approved before they become actions.
  1. 01
    Business Data
    CRM · sales · docs
  2. 02
    AI Engine
    retrieval · reasoning
  3. 03
    Analysis
    structured output
  4. 04
    Recommendation
    with evidence
  5. 05
    Human Approval
    review · edit · reject
  6. 06
    Action
    executed · logged

How we approach data & security

Responsible engineering practices as standard.

Business systems and integrations move sensitive data. We treat them as production systems from the first line of code.

We do not claim compliance certifications (such as ISO 27001, SOC 2 or PCI DSS) unless they have been independently verified. If your project requires certified infrastructure, we design around providers that hold the relevant certification.

  • Secrets and API credentials kept on the server, never in client code
  • Encrypted transport for every integration and user session
  • Role-based, least-privilege access to systems and data
  • Input validation and webhook signature verification
  • Authentication and scoped tokens for every API
  • Logging and audit trails for data exchange and approvals

Technology

The stack we work with.

Frontend

  • React
  • Next.js
  • TypeScript
  • Tailwind CSS
  • Progressive Web Apps

Backend

  • Node.js
  • Next.js Server
  • REST APIs
  • Webhooks
  • Scheduled Jobs

Database

  • PostgreSQL
  • Prisma ORM
  • Data Warehousing
  • Caching

Cloud & Delivery

  • Vercel
  • Linux Servers
  • CI/CD Pipelines
  • Monitoring & Logging

AI

  • LLM APIs (Claude and others)
  • AI Agents & Tool Use
  • Retrieval (RAG)
  • Structured Outputs
  • Human-in-the-loop Approval

Integrations

  • Meta Marketing API
  • Meta Conversions API
  • WhatsApp
  • Google APIs
  • Payment Gateways
  • CRM & ERP
  • Accounting
  • Email & SMS

Let's discuss what your business needs to build, connect or measure.

Tell us about the problem you are trying to solve. We'll respond with an honest assessment of the options — custom build, integration, automation or analytics — and a proposed first step.

Discuss Your Project

No obligation. We reply to project enquiries within one business day.