-success?style=for-the-badge){kind=icon}
I bridge telecom-grade infrastructure reliability with modern MLOps and AI deployment — building systems that run at sea, in the cloud, and at scale.
12 yrs Telecom/Network → AWS Cloud Infra → AI/MLOps Pipeline
(VSAT, VPN, WAN) (EC2, S3, IAM) (SageMaker, YOLOv8)
Currently leading AI infrastructure and data pipeline for Marine-Drone Co., Ltd (Korea):
- 🐟 19–32 TB Tuna Detection Dataset Pipeline — largest in Southeast Asia
- 🚁 51-vessel drone fleet RTSP/MQTT command-and-control infrastructure
- ☁️ Full AWS IaC via Terraform — S3, SageMaker, EC2 GPU, CodePipeline
🐟 End-to-end MLOps pipeline for the largest private tuna fish school detection dataset in Southeast Asia.
51 Vessels → RTSP Recording → EC2 MediaMTX → S3 (19–32 TB)
→ CVAT Labeling → SageMaker GPU Training → Model Registry
→ CodePipeline CI/CD → EC2 GPU Inference → MDT Gallery App
Key achievements:
- ☁️ Full IaC: S3, IAM (
MDTGallery_Datasetgroup), EC2, SageMaker, CloudWatch —terraform applydeploys everything - 💰 Cost-optimized from
$1,454/mo → **$870/mo** with Spot Instances + Intelligent Tiering - 🔐 Least-privilege IAM with access to 9 existing + 1 new S3 dataset buckets
- 📊 Real-world data: 3 years × 51 vessels × 700–1,200 flight hrs/vessel
| Repo | Description | Stack | Status |
|---|---|---|---|
| drone-station-server | Multi-vessel MQTT + RTSP fleet management | Python · Mosquitto · MediaMTX · Docker | 🟢 Public |
| drone-streaming | RTMP + RTSP dual-protocol streaming | Docker · Node Media Server | 🟢 Public |
| tuna_v2-ai-server | YOLOv8 real-time tuna detection + RTSP | Python · YOLOv8 · EC2 GPU | 🔒 Private |
| janus-streaming-server | WebRTC gateway for live drone video | Janus · WebRTC · Docker | 🔒 Private |
2009 B.Eng Electronics & Communication — Technological University Myanmar
│
2016 Telecom Executive Engineer — Petronas Carigali (HF/VHF, Satellite, PABX)
│
2017 Network & System Engineer — Golden TMH Telecom (VSAT, VPN, WAN, 99.99% uptime)
│
2024 Cloud Engineer — Nex4 ICT (KBZ Bank, AWS security, Ansible automation)
│
2025 ── Junior Cloud/DevOps — Video Streaming Platform (100+ servers, CloudFront, CI/CD)
│
2025 Cloud / DevOps Engineer — Marine-Drone Co. Ltd Korea (AI, MLOps, Drone Fleet)
│
2026 🎯 Building: 19–32 TB AI Data Pipeline · Terraform IaC · SageMaker MLOps
| Domain | What I Deliver |
|---|---|
| MLOps & AI Pipeline | Data ingestion → labeling → GPU training → CI/CD deploy → monitoring |
| Computer Vision | YOLOv8 / Mask R-CNN detection systems on custom datasets |
| AWS Infrastructure (IaC) | Terraform plug-and-play: S3, EC2, SageMaker, VPC, IAM, CloudWatch |
| Video Streaming | RTSP, RTMP, WebRTC systems — low-latency, multi-source |
| IoT / Drone Fleet | MQTT command-and-control, real-time telemetry, multi-vessel management |
| Cloud Security | IAM least-privilege, VPC, WAF, ACM, audit logging |
| Network Engineering | Enterprise VPN, VSAT, WAN, firewall design |
🚀 Available for remote consulting, contract work, and full-time roles worldwide. Specializing in: MLOps · AI Infrastructure · AWS Cloud · Drone/IoT Systems
| 📞 | +95 998 453 1435 |
| 💼 | linkedin.com/in/hnin-pwint-cho-35152a47 |
| 📍 | Yangon, Myanmar — available remote globally |
14+ years of infrastructure. From satellites to AI models. From the sea to the cloud. 🌊🚁🤖