Portfolio
KOEN

Dongho Kang

강동호

Software engineer focused on problem-solving and the completeness of services without limiting tech stack. Actively contributes from the planning stage to prioritize user convenience. Currently responsible for the overall billing and user-side services across multiple MSA environments.

(+82) 010-6590-3302 gongdongho12@gmail.com github.com/gongdongho12 linkedin.com/in/dongholab

Experience

LG Uplus VoltUp

(Tech Innovation Tribe, Back-end)
Sep 2024 — Present

Billing — Payment & arrears reliability (DLQ, recovery batches) Promotions — Coupon & point platform for user retention User-Side — Vehicle lookup, PnC auth, backend & Flutter app Internal AI — Agent routing & automated log diagnosis DX Infra — CI/CD integration & auto-doc tooling

Kakao Style

(Pricing System, Back-end)
Jun 2022 — Sep 2024

Final pricing service, product matching, Zigzag membership new development and migration, and shopping EP integrations (Naver Shopping / Google Shopping). Pipelines for external price collection and competitive position automation.

Storelink

(R&D Engineer)
Mar 2021 — Jun 2022

Product-data collection and keyword-ranking analysis platform Developed AdLink, Uzzit, Pugshop, and Friday

Whatap Labs

(Front-end Engineer)
Mar 2019 — Mar 2021

Frontend development for monitoring dashboards across infrastructure, JVM applications, and Kubernetes.

Hamschout

(SW Engineer)
Aug 2017 — Aug 2018

Developed the Itfluencer platform matching influencers with businesses.

OpenBridge

(SW Engineer)
May 2016 — Jul 2017

Developed AlarmTalk (alarm + chat), Potion (photo editing + missions), and the OpenBridge web platform.

Projects — LGU+ VoltUp

Led payment architecture from scratch, expanding into a modular multi-vendor PG structure and a multi-tiered arrears recovery system for state consistency and automated failover.

  • [Multi-Vendor Architecture] Designed vendor strategy pattern to integrate KakaoPay, TossPayments, and KakaoT with a unified interface
  • [DLQ Arrears Recovery] Built GCP Pub/Sub DLQ pattern and retry/recovery batches to keep arrears processing reliable and traceable
  • [State Transition Integrity] Hardened failover transitions and AlimTalk cancellation contexts for full, partial, and roaming payment scenarios

Outcomes

  • Full KakaoT partner integration: card registration, payment, cancellation, and promotion lifecycle implemented

Promotion Platform (Coupon & Point)

Coupon: Jul 2025 / Points: Aug 2025 — Present

Built a scalable coupon & point backend for B2B partners (Nexen, Toyota, Bluemembers) with coupon-pack payment restrictions across issuance, lookup, usage, and admin workflows.

  • [Point Wallet System] Designed point wallet system with per-deposit expiration and 7-day expiry-soon tracking
  • [3-Phase Lifecycle] Implemented 3-phase point lifecycle (Hold→Confirm→Release) ensuring automatic rollback on payment failure
  • [Payment Restriction Policy] Modeled allowed payment vendor restrictions per coupon pack across issuance, lookup, usage, and Admin creation flows

U+ VIP Coupon

Jun 2026 — Present

Integrated feapp, coupon-service, and admin-fe for monthly U+ VIP coupon issuance and created operational proxy tooling to resolve VoC bottlenecks.

  • Added UPLUS_VIP coupon pack type and benefit-month overlap prevention logic to coupon-service
  • Built member card verification, monthly issuance caps, and LGU+ approval cancellation compensation in feapp
  • Implemented Admin UI & APIs for phone-based card lookups, identity pre-checks, and instant/scheduled SMS/AlimTalk dispatch

Outcomes

  • Connected external LGU+ membership approval and internal coupon issuance end-to-end for consistent verification
  • Shifted manual developer log/policy checks to Admin pre-check screens, accelerating VoC resolution

User-Side Backend

Jul 2025 — Present

Developed core user-side backend features including KakaoT account/payment integration across 4 MSAs, PnC (Plug & Charge) auto-charge, and unpaid-charge monitoring.

  • Designed KakaoT account & payment integration across 4 MSAs (OAuth2 linking, card registration, payment approval/cancellation)
  • Built vehicle registration and PnC (Plug & Charge) charger-vehicle unique key auto-mapping & authorization service
  • Built homepage statistics API with local/remote cache and distributed locks for high-concurrency public traffic
  • Applied circuit breaker pattern to external certificate service for auto-disable and emergency recovery

Outcomes

  • Full KakaoT partner integration: card registration, payment, cancellation, and promotion lifecycle implemented across 4 MSA services

Led Flutter hybrid delivery for 2.0; designed WebView-native bridge flows and core app paths such as QR scanning, permissions, push notifications, and forced updates.

  • Rapid Android/iOS hybrid delivery via Flutter for 2.0 launch (2025.03)
  • JSBridge for frontend–hardware and app message communication
  • Built WebView-based UX flows for new-window handling, QR scanning, camera permission, and forced updates
  • Custom ML Kit-based QR scanner page and responsive scan UI
  • FCM-based push notification system
  • Crashlytics-driven Dart/native error collection and camera/FCM stabilization

Outcomes

  • Launched within 2 months (Dec 2024 — Feb 2025)
  • designed protocols for frontend to invoke native functions via JS Bridge
  • Improved core app entry points through Crashlytics-based stabilization and QR/permission/update-flow refinements

Built a Chrome Extension that recreates app-dependent flows in the browser to reduce app-attachment validation delays, then extends into replayable single-API operational corrections not supported by Admin.

  • Recreated app-supported flows such as new windows, QR scanning, camera permission, and forced-update version branches in a browser extension without connecting the app
  • Built API capture/replay to shorten app-development validation time and automate repeated QA flows
  • Extended the tool into row-based Bulk Replay for single-API corrections not directly supported by Admin, such as charge-zone correction cases

Outcomes

  • Reduced waiting and repeated interaction time by making app-dependent validation quickly testable from the browser
  • Turned one-off JS fetch operational corrections into a reusable internal tool flow

Developed Voltbot, an internal AI agent platform featuring intent-based agent routing (Voltbot Crew) and automated log diagnosis cross-checking GCP logs, guides, and GitHub code.

  • [Agent Routing] Implemented AgentRouter and AUTO_ROUTING to dynamically route user requests to specialist agents based on intent and permissions
  • [Gemini Native Function Calling & Safety Engine] Integrated Gemini Native Function Calling with a unified Agent/Tool schema conversion layer and user-approval safety controls (ToolWithApproval) over WebSocket streaming
  • [Automated Log Diagnosis] Built LogDiagnosisAgent with GCP Cloud Logging and correlation key pivoting (traceId, user_id, order_number) for multi-service triage

Outcomes

  • Reduced developer log-triage bottlenecks while standardising incident response
  • Created a single entry point automatically connecting users to the right specialist agent

Redesigned card-state management for the Ministry of Climate, Energy and Environment public roaming integration and added retry/monthly resync paths so baseline data can recover after missed events or transient failures.

  • Redesigned card-state updates in the Ministry of Climate, Energy and Environment public roaming flow from payment-response-based to billing-arrears-event-driven — selective updates only for arrears events
  • Simplified the roaming card-state processing path and consolidated billing lookups to improve throughput
  • Built an MCEE API retry process that prioritizes critical data (membership cards) over tolerably-lossy items (charger status)
  • Added a monthly full resync scheduler and task seed for MCEE member cards to prevent long-term drift from the external system

Outcomes

  • Improved data accuracy by rearchitecting card state update logic
  • Reduced processing overhead by eliminating serialization and consolidating billing queries
  • Added an operational safety net that periodically restores member-card baseline data after missed events or transient failures

AI-first reviews, mobile CI/CD hardening, remote Gradle cache, iOS TestFlight automation, obfuscation policy, Firebase distribution, and internal Admin API standards.

  • Built a reusable skill repository of microservice conventions and recurring task shapes so it could be applied across generative LLM tools such as Claude Code, Cursor, and Gemini
  • Gemini CLI for first-pass review load shedding
  • Docusaurus + GitHub Actions auto-doc system with git-polling change detection
  • Firebase App Distribution for faster QA and release cycles
  • Gradle generator for application-local.yaml
  • Standardized Admin internal API boundaries with admin-internal-* clients and an X-Internal-Caller header convention
  • Simplified Android/iOS app delivery to reduce key-management burden and avoidable failure points
  • Refactored Android Jenkins into shared libs; GCS remote cache; iOS GA/TestFlight pipeline; 300+ line R8 rules

Outcomes

  • Reduced review overhead via automated first-pass AI review
  • Shortened QA/deploy cycle and improved QA self-serve with Firebase Distribution
  • Auto-doc system reduced manual documentation maintenance
  • Improved release stability by reducing mobile delivery key-management burden and avoidable failure points

Projects — Kakao Style (2022.06 — 2024.09)

Mobile IP Pool Proxy Infrastructure & Dynamic Pricing

Jan 2024 — Sep 2024

Built a mobile IP pool proxy infrastructure to bypass crawling blocks and automated a dynamic pricing pipeline based on price competitiveness.

  • Allocated per-port mobile HTTP proxies and developed proxy pool lookup APIs
  • Built mobile device control APIs (airplane mode, reboot) and automated rate limit management
  • Constructed matching datasets between external (EP, lowest price) and internal products via AWS Athena
  • Developed batch jobs for dynamic pricing based on price-gap analysis and Excel report generation

Overhauled 1:1 matching to an N:M structure to match identical products across all stores and types.

  • Implemented dual caching for full matching lookups and served "Winner" data for search
  • Developed matching batch jobs using AWS Athena for items with ≥99.5% image similarity
  • Built matching lookup APIs for review integration and automated matching via Kafka consumers

Unified scattered pricing logic into a single API so coupons, promotions, and other benefits follow one contract.

  • Applied timeouts and circuit breakers for service stability
  • Consolidated scattered pricing logic into a unified API across company-wide gateways and ensured price exposure consistency

Membership & Mileage Migration

Apr 2023 — Jun 2023

Redesigned membership tiers for retention and migrated the legacy membership service to Spring Boot.

  • Zero-downtime migration of the legacy membership service (cormo.js-based) to Spring Boot via 1:1 DB migration
  • Expanded the membership tier calculation period from 3 to 6 months
  • Optimized 3-month tier calculation by caching monthly totals and using cumulative sums

Outcomes

  • Reduced DB load from 70% to under 30% through strategic caching
  • Achieved zero-downtime deployment through API I/O integration tests and gradual gateway switching

Enhanced price competitiveness of Zigzag products for exposure on Naver Shopping price comparison.

  • Developed batch jobs for GMV-competitive products and integrated card promos/delivery badges
  • Implemented logic to compare prices with competitors and consumed real-time Kafka updates
  • Commonized the Engine Page and lowest-price update flow so it could be quickly extended to Google Engine Page (YouTube Shopping), while also building backoffice tooling and leading the CPC-to-CPS transition

Outcomes

  • To satisfy the 2-hour product update interval required for CPS, replaced full-catalog refreshes with partial updates for sold-out or price-changed items, enabling generation of the Naver Shopping feed CSV and changed-product sync dataset within an hour
  • Built a shared Engine Page and lowest-price update structure that could be extended quickly to Google Engine Page (YouTube Shopping)
  • Created the path to consolidate team-specific final-price APIs into a unified Final Pricing API, aligning previously inconsistent pricing logic across surfaces

Projects — Earlier experience

Friday analytics

Mar 2021 — Jun 2022

Backend development for commerce keyword analytics service.

  • Introduced Kotlin and WebFlux as the main tech stack for the entire team
  • Replaced core service features with Friday search APIs (still in use)
  • In-house automation tools: Jenkins CI/CD, WireGuard VPN, NPM modules/NPX boilerplates
  • Cross-architecture Docker builds (AMD64/ARM64) via buildx for AWS Graviton2 compatibility
  • Puppeteer crawler API — page access, API query, tag selection, and JS function injection

Whatap monitoring FE

Mar 2019 — Mar 2021

Developed integrated monitoring dashboards to visualize data collected from various agents (App, Server, DB, Container).

  • Developed core components including TimeSelector and Server List dashboards
  • Integrated legacy page handlers on React components and developed agent setup/install screens
  • Implemented common data processing methods for better code reusability

Outcomes

  • Reduced API polling churn for real-time dashboards and mitigated memory pressure on chart-heavy screens
  • Used the product from an operator's perspective to shape meaningful metric/chart combinations and dashboard compositions for troubleshooting and performance analysis

Influencer platform

Aug 2017 — Aug 2018

Developed an influencer marketing platform for campaign management and influencer matching.

  • Developed influencer/blogger review and verification systems
  • Built Naver search ranking lookup APIs
  • Developed advertising reporting services for clients

Outcomes

  • Automated advertiser–influencer matching, reducing contact overhead
  • Improved marketing accessibility for small businesses and startups

AlarmTalk (Android)

May 2016 — Aug 2017

Social alarm service combining alarms and chat for mutual wake-up calls.

  • Led the development of AlarmTalk 2.0
  • Implemented voice recognition and shake-to-wake alarm features
  • Built the in-app alarm store system

Potion (Android)

May 2016 — Aug 2017

Mission-based photo filter app where users earn points for printing photos.

  • Developed core photo adjustment components (brightness, contrast, crop, etc.)
  • Implemented photo filter engine and effects through design collaboration
  • Integrated account systems and mission-related APIs

Side Projects

Clip Atlas (clip-indexer) — Media Tag Generator Personal Project

Ongoing

Built a CLI & local Web media manager that scans media files, combining FFmpeg and LLM multimodal scene/voice analysis for smart file renaming, tag sidecar storage, and web UI file management.

  • [Media Indexing Engine] Developed high-performance media indexing pipeline, CLI tools, and HTTP web server
  • [Multimodal Analysis] Built scene/voice tagging algorithms combining FFmpeg frame extraction and LLM multimodal analysis
  • [Local Web File Manager] Created interactive web UI for folder planning, metadata sidecar generation, and dry-run file movement

Commit Map Personal Project

Ongoing

A map-based travel planning service built to share my travel plans and quickly draft route ideas.

  • Built as a personal service that shows travel plans and routes together on a map and timeline
  • Added an AI workflow that turns natural-language destinations and routes into a rough itinerary draft, which can then be refined directly in Markdown
  • Manages trip content through Markdown-based travel data and static deployment so a single link can be used to share plans easily

Teddy Resume Personal Project

Ongoing

A personal resume service that manages multilingual resumes and project portfolio pages from a single YAML source across multiple layouts, optimized for print and PDF output.

  • Structured the app so Korean/English resumes, project portfolio pages, and multiple resume layouts are managed from one YAML data source
  • Implemented output UX for real submission and sharing scenarios, including A4 auto-pagination, no-margin print guidance, background-graphics hints, and PDF export flow
  • Continuously refactored shared components and content structure so project links, common language switching, and repeated wording/layout tuning can be applied quickly

Other Activities

Kookmin SW-Centered University Mentor Mar 2024 — Present

Mentored students on SW architecture, problem-solving, and user flow optimization

[Mentor Profile]
Codeit Backend Course Mock Interview Mentor Sep, Dec 2025

Technical and mock interview mentoring for backend bootcamp graduates

Kernel360 Backend Mentor (1st ~ 5th cohorts) Dec 2023 — Aug 2025

Guided collaboration-based enterprise problem-solving projects across 5 cohorts

Kakao Tech Campus Backend Mentor (2nd) Jun 2024 — Aug 2024

Mentored backend track project teams for 2nd Kakao Tech Campus

Hanyang ERICA SW Startup Lecture Feb 2024

Gave a lecture on "Effective Backend" for SW startup series

Boostcamp 8th Backend Mentor Nov 2023 — Dec 2023

Mentored two iOS service project teams for Boostcamp

Hanbit N MSA — Monitoring session Dec 2023

Led the 9th session on MSA Monitoring for Hanbit Media seminar

Boostcamp 7th~8th Backend Reviewer Sep 2022 — Oct 2023

Conducted campus meetings and code reviews for backend campers

UbuCon Korea 2023 Speaker Sep 2023

Presented on "Building a Development NAS using WSL"

SW University Joint Hackathon 2023 Mentor Jun 2023

Provided FE/BE technical mentoring and career counseling

[Mentor Profile]
Programmers School JAVA Mentor (11th) Mar 2023 — Apr 2023

Provided Q&A, code reviews, and video mentoring

Gyeongbuk SW High School React Lecture Sep 2022 — Oct 2022

Conducted React lectures and project guidance for high school students

Ubuntu Korea Community Infra Manager Mar 2018 — May 2021

Managed community infrastructure including Forum, Wiki, and mail servers

AngelHack 2020 Winner (Common Computer Prize) 2020

Developed local group-buying platform LiBi; awarded Common Computer Prize

[Related]