Work

Case studies covering client delivery and ML/edge product work, ordered by business impact priority.

All case studies

Enterprise CRM - Sales Pipeline & Performance System visual

Client Delivery

Enterprise CRM - Sales Pipeline & Performance System

  • Problem

    Sales pipeline updates and performance reporting lived across disconnected files, causing slow reviews and unclear ownership.

  • What I did

    Mapped sales stages and reporting logic before schema design

  • Result

    Centralized pipeline and performance reviews into one live dashboard used for weekly decision cycles.

Designed and delivered a custom CRM that centralized pipeline tracking and performance reporting into one operational workflow.

  • Laravel
  • Next.js
  • PostgreSQL
  • Docker
  • Nginx
Read case study
Media Android App - AI Chat + Search UX visual

Client Delivery

Media Android App - AI Chat + Search UX

  • Problem

    Users lacked a clear transition from AI chat prompts to meaningful search actions, and state sync issues reduced UX reliability.

  • What I did

    Introduced WebView AI chat entry aligned with existing navigation

  • Result

    Shipped a clearer AI-assisted discovery flow with fewer state mismatch edge cases.

Improved AI chat entry and search behavior with suggestion bubbles and state-handling fixes in a live Android app flow.

  • Android
  • Kotlin
  • WebView
  • Search UX
Read case study
Automation Workflows visual

Client Delivery

Automation Workflows

  • Problem

    Inbound orders and chat updates were copied manually into spreadsheets, producing inconsistent records and delayed follow-up.

  • What I did

    Implemented web-order-to-sheet automation with validation and normalization

  • Result

    Replaced repetitive manual logging with automated lead/order pipelines and cleaner operational visibility.

Built automation pipelines (n8n, GAS, APIs) for web order ingestion and WhatsApp logging/tagging into a shared operations sheet.

  • n8n
  • Google Apps Script
  • REST APIs
  • Google Sheets
Read case study

Client Delivery

Multi-Surface Platform Monorepo

  • Problem

    Separate repositories for related surfaces caused dependency drift and slower coordinated releases.

  • What I did

    Structured repository with npm workspaces

  • Result

    Reduced release friction by aligning frontend and backend delivery in one repository.

Consolidated admin, landing, and API development in a single monorepo with npm workspaces.

  • npm workspaces
  • Next.js
  • NestJS
  • TypeScript
Read case study

Client Delivery

Social Feed UI Fixes and Card Polish

  • Problem

    Social cards had uneven spacing and hierarchy, reducing clarity in dense feed views.

  • What I did

    Reviewed spacing rhythm, card hierarchy, and interaction cues

  • Result

    Improved feed scannability and visual consistency without broad redesign risk.

Delivered targeted UI polish and bug fixes on social feed cards to improve readability and consistency.

  • Frontend UI
  • Responsive CSS
  • Design QA
Read case study

ML / Edge

I-ScanTea

  • Problem

    Manual quality checks required repeated visual comparison and lacked a fast digital workflow.

  • What I did

    Prepared dataset and labeling workflow for target classes

  • Result

    Demonstrated on-device classification workflow for practical field capture scenarios.

Built an edge ML prototype for tea leaf quality scanning with mobile-first inference flow.

  • TensorFlow Lite
  • Android
  • Python
  • Edge ML
Read case study

ML / Edge

Scanberry

  • Problem

    The workflow needed rapid classification support in environments with unstable connectivity.

  • What I did

    Built lightweight model iteration and evaluation loop

  • Result

    Validated a practical on-device scan flow for classification without cloud dependency.

Created a mobile computer-vision prototype for fruit scanning workflows with edge inference constraints.

  • Computer Vision
  • Android
  • TensorFlow Lite
  • Mobile UX
Read case study

ML / Edge

OffScan AI

  • Problem

    Users needed OCR extraction with stronger privacy guarantees and no network dependency.

  • What I did

    Designed offline-first OCR flow for camera/image import

  • Result

    Published an app workflow that performs OCR locally without cloud upload requirements.

Built and shipped an offline OCR Android app focused on privacy-preserving, on-device text extraction.

  • Android
  • On-device OCR
  • Edge AI
  • Mobile Product
Read case study