GameDuo

Resolved delayed balance issue response caused by inability to track data change history during game operations

• Inability to track data change history across environments and projects — Designed Git-like version control system with UUID-based cross-environment/project entity tracking, Ensured data consistency across 6 games
• Manual entity change recording prone to omission — Applied Event Sourcing-based change tracking with Auditable decorator + TypeORM Subscriber pattern, Automated entity change recording pipeline
• Multi-environment version conflicts during data merge — Developed 3-Way Merge Engine with parent/child entity conflict detection and unique constraint handling, Enabled reliable version merging across 6 games
• Expensive full-snapshot comparison for every version diff — Designed Version Diff Engine with dual strategy: incremental comparison and snapshot comparison based on Base Audit availability, Established cost-efficient version diffing
• Entity tracking failures due to PK dependency during migration and merge — Introduced shared identifier-based entity tracking decoupled from PK dependency, Ensured accurate entity tracking during migration, comparison, and merge

Unified marketing platform managing Google Ads/Meta/TikTok campaigns, creatives, and metrics in a single system, built on dedicated marketing DB separation

• Fragmented ad platform management across Google Ads/Meta/TikTok — Delivered unified campaign automation API for creation, deployment, modification, and retention metrics in a single platform, Consolidated multi-platform campaign management into one system
• NAS-S3 sync reliability and restore failures — Separated SQS server-worker flow, automated outbox events, tracked asset_nas_sync state, Secured restore reliability with state-tracked sync pipeline
• Meta asset sync performance bottleneck — Converted ORM single-row saves to bulk processing, Image sync reduced 72% (25.7s→7.1s), DB transactions reduced 95% (10~16s→0.3~0.5s)
• High creative processing Lambda costs on x86 architecture — Migrated Lambda to ARM64 Graviton2 architecture, Reduced creative processing Lambda costs by 20%
• High BigQuery query costs for marketing data reads — Adopted BigQuery Storage Read API with gRPC streaming-based high-performance data reading, Reduced BigQuery costs by 82% ($6.25→$1.1 per TB)
• Marketing query latency (18s) from 100+ column denormalized table — Normalized table into main/time-series/prediction, tuned indexes + cursor pagination, Performance improved 97% (18s→0.5s)
• The Asset table view suffered severe jank — 30 rows × 13 columns (390 cells) re-rendering on every dropdown open or row selection, combined with an infinite useEffect render loop — Applied React.memo with custom comparator to TableRow, extracted per-row cells with isolated hook subscriptions, reduced columns useMemo dependencies from 12 to 2, and guarded controlled-mode useEffect to break the loop, Eliminated unnecessary re-renders and resolved the infinite render loop; shared Table component propagated the fix to 36 downstream tables
• Marketing RCP data was only viewable in spreadsheets, and the existing BigQuery aggregation table combined multiple sources — creating data source divergence that blocked unified console queries — Switched the BigQuery source table from the multi-source aggregate to a single-source table, introduced Criteria-Result pattern field-metadata/filter-options APIs, and implemented a FE multi-select filter with a TOTAL row aggregation, Structurally resolved data source divergence via single-source migration, embedded unified RCP queries into the console, and replaced hardcoded currency/decimal display with BE metadata-driven formatting

Resolved batch job limitations from marketing metrics 60-day → 360-day expansion and serverless transition for batch processing

• Full serverless migration risked operational stability — Designed hybrid architecture keeping API server on EC2 while separating batch/job processing to Lambda, Attained workload-optimized resource usage without disrupting live services
• Batch processing limited to 60-day collection range with 2-hour runtime — Established Event-Driven flow with SQS+Lambda+EventBridge, Reduced batch time 2h→5min, expanded collection 6x (60→360 days)
• Data consistency risk during event publishing — Adopted Transactional Outbox Pattern for scheduled and delayed event publishing, Ensured data consistency across distributed event processing
• Lambda throttling, high log costs, and build OOM issues — Applied Batch Size bulk processing, CloudWatch log optimization, and build OOM remediation, Stabilized Lambda operations and reduced operational costs
• DB connection exhaustion during massive Lambda execution — Introduced RDS Proxy connection pooling, Resolved connection exhaustion and stabilized database access

Built S3-based sync and automated DDL management system to resolve dynamic game data inconsistency across environments

• Dynamic game data inconsistency across environments — Built S3-based cross-environment data synchronization across development/staging/production, Unified game data state across all environments
• Manual DDL schema management causing sync failures — Created automated DDL management engine with dynamic PK column type resolution, column type mismatch detection with MODIFY, and automatic index creation/RENAME, Automated schema sync across environments
• Large-scale Cloud Data ingestion and S3 upload bottlenecks — Analyzed and optimized ingestion and upload pipeline, Resolved large-scale processing delays
• Sync job instability causing operational issues — Implemented job separation, transitioned scheduling approach, tuned timeouts, and introduced non-blocking processing, Secured operational reliability for sync pipeline
• Four CloudData bugs: Redis cache invalidation, DDL SKIP metadata, copy key deletion, excludeCloudData propagation — Diagnosed and fixed all four issues systematically, Stabilized CloudData operations
• Unintended full Cloud Data deletion risk from missing option forwarding — Added excludeCloudData option forwarding to four POST migration paths, Blocked unintended data deletion risk

Development and operation of NestJS utility (10 modules) + Game Server Kit (2 packages) for multi-project code consistency

• Common code duplication across services increasing maintenance cost — Designed 10-module library system: core, repository, cache, lock, slack, crypto, smb, hash, type, iac, Unified shared code across 7 projects with standardized dependency management
• Repository module lacking bulk operations and audit logging capabilities — Enhanced Repository module with Bulk/Audit Log/TypeORM narrowing using overloading + TypeScript generics; refactored 2000+ lines by SRP, Improved module extensibility and maintainability
• Concurrency conflicts in multi-instance environments — Engineered distributed lock module with ElastiCache (Redis)-based distributed lock decorator, Enabled safe concurrency control across multi-instance deployments
• Manual library upgrades taking 3 hours across 7 projects — Added workflow_dispatch+matrix and changed-package CI tests on GitHub Packages, Reduced 7-project upgrade runtime 3h→15min
• Game server shared code tightly coupled in monolithic package — Extracted sheet processing module + 5 submodules into independent package, resolved conflicts via 5-branch divergence analysis, Realized independent package versioning and deployment
• Slow CI pipeline (15m47s) due to test infrastructure inefficiency — Switched ts-jest isolatedModules and explicit Entity types, Cut CI time 61% (15m47s→6m06s); passed 81 suites/978 tests

Redesigned the in-game notification translation domain model — decoupling AI translation from library-based conversion into a 3-stage pipeline

• Simplified Chinese was modeled inside the AI translation settings domain despite not being AI-supported, and settings storage/retrieval depended on the Traditional Chinese AI translation state — making independent toggling impossible without a breaking change — Extracted the derived-conversion module from the AI translation module into 9 use cases with dedicated entity/repository/scheduler, and applied the global-inheritance pattern for effective settings, Clarified domain boundaries and enabled independent toggling of derived conversion rules
• Decoupling required a breaking API change — Simultaneously deployed FE/BE/Gateway with role permission migration for zero-downtime transition, Completed the breaking API change without incident and cut effective-settings DB queries by 50% (4→2)
• The detect cron lacked version priority, delaying translation of the latest version; 30-second interval further slowed response — Added version ID-based priority ordering to detect SQL, shortened the processing interval from 30s to 10s, and raised batch limit from 100 to 200, Eliminated latest-version translation lag; progress UI plus total-count caching improved user visibility

Built probability calculation package and CDK-based audit log analytics infrastructure to address regulatory risk from lack of game probability verification

• No reusable probability calculation module across game projects — Packaged NestJS DynamicModule with 5 probability functions + Kinesis logging as shared probability package, Integrated into 12 games with standardized probability calculation
• No audit log analytics infrastructure for probability verification — Codified CDK analytics pipeline: Kinesis→Firehose (Dynamic Partitioning+Parquet)→S3→Glue→Athena, Unlocked end-to-end probability audit log querying
• Mock-based tests lacking regression confidence for infrastructure code — Replaced mocks with LocalStack + Testcontainers integration tests, Secured 94%+ coverage (12 suites/115 tests)

Games on AWS 2025 Customer Session

Shared a case study of a single engineer scaling data pipeline capacity 270x in 10 days, measured by hourly concurrent job throughput