- Based in Ho Chi Minh, Vietnam
- Backend Engineer maintaining Sendo's post-revamp WMS platform across warehouse, inventory, order allocation, export, and e-invoice services
- Built production Rust services: e-invoice processing ~15K invoices/day and streaming exports up to 6M rows with ~128MB peak RAM
- Focused on systems-oriented Rust, event-driven backend design, and production observability
- 🏆 Outstanding Employee of 2025 at Sendo
- Rebuilt order allocation into an event-driven flow: ~90% faster, with 200 orders allocated in 5s
- Led a Rust export revamp that reduced memory peak from ~5GB to ~128MB for exports up to 6M rows
- Resolved N+1 bottlenecks across warehouse modules: ~30s → <200ms
- Built a Rust-based e-invoice service processing ~15K invoices/day in production
- Designed and built the core inventory service with sub-150ms location lookups
- Automated purchase order generation and reduced creation time by ~60%
| Project | Description |
|---|---|
| Export Flow Revamp (Rust) | Technical lead (3-dev team). Streaming export pipeline replacing a legacy Go service; ~5GB → ~128MB RAM for exports up to 6M rows |
| E-invoice Service (Rust) | Technical lead (3-dev team). Delivery-to-invoice automation pipeline processing ~15K invoices/day in production |
| Purchase Order Automation | Sole implementer. Rule-based daily PO + invoice generation; ~60% time reduction vs manual |
| Warehouse Revamp | Contributed to splitting a shared warehouse platform into 10+ .NET microservices and SQL Server → PostgreSQL migration |
| Order Allocation | Replaced synchronous flow with event-driven processing; >90% faster, 200 orders in 5s |
| Inventory Service | Designed and built from scratch with concurrency safeguards and sub-150ms location lookups |
| Project | Description |
|---|---|
| Game Server (.NET 10 + Rust) | Self-hosted multiplayer game platform with a TCP server, admin CMS (Go + React), Rust analytics pipeline, mail sender, and K3s-based infrastructure |
| rust-rabbit | Open-source RabbitMQ client for Rust with MassTransit-compatible interoperability, retry strategies, and DLQ support. Published on crates.io |
🏗️ Architecture Principles I Follow
- Event-Driven Architecture with message brokers (RabbitMQ / Kafka / NATS)
- Clean Architecture with clear domain boundaries
- Read/write separation where high-throughput workflows require it
- Containerized deployments with Docker
- Observability-first: structured logging, distributed tracing, metrics dashboards (Grafana, Loki, Graylog)
