William Lo Channiko

Applied AI + Full-Stack Engineer

William Lo Channiko

I build AI systems for evidence-heavy work, with software-enforced rules and explicit human decisions.

View projectsDownload resume
William Lo Channiko

AI should support the decision, not hide it.

Models contribute inside a workflow whose context, rules, and decision authority stay visible.

  1. ContextSource material stays visible
  2. AI proposalBounded, reviewable contribution
  3. Software checksDeterministic rules own constraints
  4. Human decisionAuthority remains explicit

Selected work

Full-stack AI systems, built end to end.

Each case study covers the product problem, architecture, control boundary, verified engineering results, and what remains outside the current scope.

Invoice Review

AI-powered invoice review & approval system

Invoice Review workspace showing a synthetic invoice beside extracted fields, validation findings, and an approval blocker.
Review workspace with synthetic fixture data

ERPNext workflow benchmark complete

A human-controlled accounts-payable workflow that keeps the source PDF, AI-proposed fields, deterministic validation, reviewer corrections, and approval-gated export in one inspectable system.

68%
Lower median invoice-to-ERP draft time (153s to 49s)
10/10
Expected assisted-workflow results, versus 9/10 through direct entry
98.75%
Exact field match on a sealed 10-invoice synthetic holdout (79/80)

Current boundary

One operator used synthetic invoices in a local ERPNext sandbox. The evidence does not establish production-user adoption, customer impact, or multi-user time savings.

View case study
Repository

Case Resolution Copilot

Policy-governed decision workspace

Support escalation Decision Brief showing verified facts, missing information, a suggested resolution, uncertainty, and required human review.
Decision Brief for a synthetic support case

Controlled-pilot readiness gate passed

An authenticated workspace for assembling case evidence, applying versioned policy, preparing a reviewable resolution, recording human approval, and reconciling uncertain action outcomes.

~84%
Lower raw median case-resolution workflow time (582s to 95s)
3/3
Complete safe workflows, versus 0/3 through the manual path
375 ms
Warm authenticated primary-content readiness against a 2,500 ms gate

Current boundary

The benchmark used matched synthetic cases and one developer operator. The Gmail journey was bounded to approved draft creation with no automatic send; no production-user or customer-impact claim is made.

View case study
Repository

Professional experience

Full-stack engineering inside an operating business.

Two internship terms covering native Android, offline workflows, APIs, an internal web system, approval flows, imports, and business reporting.

PT Dover Chemical

Full-Stack Software Engineer Intern

Built internal sales-operations software for PT Dover Chemical across the CRM Dover Chemical Android app and a Customer Management System, covering offline synchronization, approval workflows, master-data management, imports, and analytics.

Jan-Aug 2025

CRM Dover Chemical - Android App

  • Implemented Draft Sales Order workflows in Kotlin for the CRM Dover Chemical Android app, comprising 8 screen classes and 38 HTTP API operations for authentication, reference data, transactions, status workflows, and file attachments.
  • Implemented offline persistence with 15 Room entities, 15 DAOs, and 82 data-access functions, plus a WorkManager sync workflow covering server-ID assignment, item submission, attachment upload, retry handling, and local cleanup.
  • Kotlin
  • Room
  • Retrofit
  • WorkManager

Aug 2025-Jan 2026

Customer Management System - Web

  • Built a Customer Management System in PHP and CodeIgniter comprising 4 application modules, 20 controllers, 17 services, and 27 application views for a customer master dataset reported at 6,656 records.
  • Implemented 90 public controller actions supporting CRUD, server-side search, filtering, sorting, pagination, file attachments, and internal JSON/AJAX workflows across 15 customer-data domains.
  • Developed role-gated approval workflows for 10 entity types with transactional updates, audit logs, reopen and reject paths, validated Excel imports, and a sales-activity dashboard with two charts.
  • PHP
  • CodeIgniter
  • MariaDB
  • Bootstrap
  • jQuery
  • Chart.js
  • PHPExcel

Engineering AI as a system, not a feature.

I am an early-career engineer based in Indonesia with full-stack experience across Android, web, APIs, and business data. My current work focuses on AI products where the model is useful without becoming the source of truth or the final authority.

Experience
1 year as a full-stack software engineer
Based in
Indonesia (UTC+7)
Education
Bachelor's Degree in Informatics, graduated with Distinction
Current focus
Document operations and support decision systems

Let’s build software people can trust.

Open to Applied AI and full-stack engineering opportunities with teams that value product clarity, measurable verification, and honest system boundaries.

Email WilliamDownload resumeLinkedInGitHub