brianreynald

Collaborated with Gojek to implement and optimize MPMD and GMV High Value Tiering endpoints, consolidating 30,000 multi-pickup orders/month and categorizing 409,000 high-value plus 2.11M non-high-value orders/month to enable trusted-driver allocation and reduce delivery costs. Developed cashless logistics system eliminating manual processes, improving efficiency by 16% and supporting 19,000 monthly orders. Automated failed COD pickup handling, reducing blocked orders by 57.4%. Created shipping fee reconciliation tool, reducing monthly losses from IDR 1.5B to 469.8M (70% improvement). Expanded COD system for new courier partners, scaling from 5k to 30k daily orders. Enhanced data privacy in APIs by masking sensitive product, resulting in 10k additional adult product orders. Optimized warehouse delivery logic to combine same warehouse source into single delivery, saving IDR 324M annually. Implemented high-value insurance tiering, securing IDR 41B in orders and generating IDR 60M revenue in 3 months. Implemented First Mile Discrimination (FMD) feature by setting seller pickup eligibility threshold, reducing pickup volume by 21% and saving IDR 418M monthly. Developed Tokopedia Content Management System for Seller Education Platform, boosting new seller activation by 5%. Developed the 'Product Recommendation Widget', providing sellers with personalized insights which boost sales performance by 7%.

Maintained critical APIs processing 100+ daily reporting for ATM cash operations with 99.99% availability. Reduced operator reporting time from 6 PM to 4 PM (33% faster) by enabling direct on-site cash deposit and automated logging of jammed currency. Secured legacy banking system by migrating from Struts to Spring Boot, eliminating critical vulnerabilities while maintaining 99.99% availability during transition. Developed OCR service for credit card applications, reducing manual data entry errors, increasing processed applications by 8.3%. Developed historical usage analytics to identify high-volume templates, optimizing 5 most-used models to reduce average OCR processing time by 12%. Developed predictive OCR quota monitoring that prevented 100% of outages while eliminating monthly manual audits. Achieved 100% GDPR compliance for OCR data by implementing RBAC with tiered access permissions. Automated monthly purging of 30+ day-old documents via cron jobs, reducing storage costs by $1.3k/year. Eliminated frontend timeouts and reduced cloud costs by $1.5k/month by implementing IBM MQ for decoupled OCR processing. Reduced redundant work by 5% by adding document checks to prevent reprocessing of identical files.