kowz@ekasithk: ~/portfolio
kowz@ekasithk:~$ whoami

██╗  ██╗ ██████╗ ██╗    ██╗███████╗
██║ ██╔╝██╔═══██╗██║    ██║╚══███╔╝
█████╔╝ ██║   ██║██║ █╗ ██║  ███╔╝
██╔═██╗ ██║   ██║██║███╗██║ ███╔╝
██║  ██╗╚██████╔╝╚███╔███╔╝███████╗
╚═╝  ╚═╝ ╚═════╝  ╚══╝╚══╝ ╚══════╝

Ekasith Kowcharoen / kowz

Civil Engineer × Digital-Twin & AI Developer × Researcher

kowz@ekasithk:~$ cat role.txt

I build AI-driven digital twins for critical infrastructure — pairing a civil-engineering foundation with full-stack software and peer-reviewed computer-vision research.

📍 Bangkok, ThailandOpen to consulting & new opportunities
Hire me Download CV GitHub
Ekasith Kowcharoen — 2nd place, Construction Innovation Challenge

🥈 2nd · Construction Innovation Challenge

● available for workkowz@ekasithk.com

kowz@ekasithk:~$ ls ~/work --quick

Explore my work

demoDigital Twin Dashboard

Interactive demo · EGAT substation monitoring

demoAirport AI

Passenger forecasting + security RAG · live demo

paperResearch Paper

ESPCN-YOLO · MDPI Buildings (2025)

wipPixel Trading Office

AI-agent trading system — coming soon

soon
wipRoadProof / $PAVE

AI-validated road-hazard network — concept demo

codeGitHub

Code, projects & experiments

kowz@ekasithk:~$ cat about.md

About

Ekasith Kowcharoen (“Kowz”) is a civil engineer specialising in construction management and a developer focused on digital twins and applied AI. He completed his M.Eng at Kasetsart University — publishing peer-reviewed research in MDPI Buildings on computer-vision PPE detection — and is now a PhD candidate continuing that work. Today he leads the platform, AI, and project-management effort on a Digital Twin study of EGAT’s high-voltage substations. Before tech, he was a national Blade & Soul esports champion (Team 200IQ), and he brings that same competitive discipline to engineering.

# languages
ThaiNative
EnglishProfessional
# status
Research & development since 2024
kowz@ekasithk:~$ skills --list --grouped

Skills

Software / Development

PythonTypeScriptFastAPINext.jsReactSQLAlchemyPostgreSQLMongoDBRedisPyTorchComputer Vision (YOLO)DockerCI/CDAWSCloudflareTerraform

Civil Engineering

Construction ManagementProject ManagementAsset-Lifecycle ManagementDigital-Twin ModellingHigh-Voltage Substation SystemsData Interoperability

Research

Computer VisionDeep Learning (YOLO / ESPCN)Image Super-ResolutionAcademic Writing & Peer ReviewData AnalysisExperimental Design

Esports / Soft Skills

Team LeadershipTeam ManagementCompetitive StrategyPerformance Under PressureCommunication
kowz@ekasithk:~$ ls -la ./projects

Projects

EGAT Digital Twin — High-Voltage Substations

Researcher · Platform, AI & Project Management

in-progress

Research project for the Electricity Generating Authority of Thailand (EGAT): “A Study on Data Interoperability Across the Asset Lifecycle with Commercial Digital Twin Platforms: A Case Study of High-Voltage Substations (DG–GEN, DG–TRN).” I lead the digital-twin platform build, AI integration, and project management/delivery.

  • ▸Interactive demo dashboard at dt-demo.ekasithk.com
  • ▸End-to-end asset-lifecycle data interoperability study
  • ▸Case study: high-voltage substations (DG–GEN, DG–TRN)
PythonFastAPINext.jsDigital TwinAI / MLPostgreSQL
Live Demo

Airport AI — Forecasting & Security Assistant

Creator · Full-stack, AI & Data

in-progress

A self-directed build exploring practical AI for airport operations. The first system forecasts daily passenger throughput (a seasonal baseline, a LightGBM model, and a flight-schedule leading indicator that compete on a walk-forward backtest, with the winner served) and turns it into an hour-by-hour checkpoint wait-time and lanes-needed plan. The second is a RAG chatbot that answers airport-security questions only from a cited document index — refusing when an answer isn't grounded — and replies in the asker's language. Modelled on public TSA throughput data at single-hub scale; provider-agnostic LLM (local Ollama for dev, Claude on AWS Bedrock for production).

  • ▸Live departures-board (FIDS) demo at airport.ekasithk.com
  • ▸Backtested forecaster that auto-serves the champion model
  • ▸Grounded, cited, multi-language RAG — refuses when unsure
PythonFastAPILightGBMRAGpgvectorNext.jsAWS Bedrock
Live Demo

RoadProof / $PAVE

Creator · Product, Tokenomics & AI

in-progress

A mobile DePIN concept on Solana. Drivers' accelerometer + GPS detect road hazards, an AI layer validates the data and filters out spoofers, and cities/insurers pay for a ranked repair map. I built a blockchain from scratch to learn the fundamentals, designed the tokenomics and threat model, and shipped an interactive concept demo. A research/build project — not a token sale.

  • ▸Interactive concept demo at pave.ekasithk.com
  • ▸AI-validated data — a 'confirmation over trust' design
  • ▸Fair-launch, utility-first tokenomics
SolanaPythonAI / MLNext.jsDePIN
Concept Site

Pixel Trading Office

Creator

in-progress

A personal R&D project — a pixel-art office where autonomous AI agents research and trade. Currently in early development; details to follow as it matures.

PythonAI Agents
kowz@ekasithk:~$ git log --work

Experience & Awards

Researcher

2024 – Present

Kasetsart University — EGAT Digital Twin Project · Bangkok, Thailand

  • ▸Lead the end-to-end build of a digital-twin platform for EGAT high-voltage substations — architecture, AI integration, and delivery.
  • ▸Designed and shipped the monitoring dashboard (public demo: dt-demo.ekasithk.com), unifying asset-lifecycle data across sources.
  • ▸Drive the study of data interoperability between full asset-lifecycle data and commercial digital-twin platforms (DG–GEN, DG–TRN case studies).
  • ▸Authored a peer-reviewed paper in MDPI Buildings (2025) on computer-vision PPE detection for construction safety.

2nd Place — Construction Innovation Challenge (Season II)

2024

Sino-Thai Engineering & Construction (STECON)

Competed solo against teams of 5–8 members, representing Kasetsart University — placing 2nd nationally. Awarded ฿100,000 (team prize) plus ฿50,000 to the university.

kowz@ekasithk:~$ ./run --achievement-mode esports

Esports — National Champion

200IQ lifting the Blade & Soul championship trophy
Team 200IQ logo

Blade & Soul · Thailand

Team 200IQ

Player & Team Manager2017 – 2019

$ cat achievements.log

  • 🏆BSTC 2017 Champion (Team Natural Selection) — swept the national grand final 5–0 (฿250,000) to represent Thailand at the World Championship.
  • 🏆Blade & Soul World Championship 2017 (South Korea) — competed among 9 global regions as the Thai national champion.
  • 🏆BSTC 2018 Champion (Team 200IQ) — defeated Goliath 6–2 (฿250,000), qualifying for the World Championship.
  • 🏆Blade & Soul World Championship 2018 — 4th place worldwide.
  • 🏆BSTC 2019 (Team QI002) — national competitor.
  • 🏆Blade & Soul World Championship 2019 — represented Thailand on the global stage.

# why it matters: competing at a national and world level forged the teamwork, strategy, and grace under pressure I bring to engineering and software delivery.

Ekasith Kowcharoen during competition day
kowz@ekasithk:~$ cat research.md

Research & Publications

Peer-reviewed · Buildings (MDPI) · 2025

ESPCN-YOLO: A High-Accuracy Framework for Personal Protective Equipment Detection Under Low-Light and Small Object Conditions

Read paper DOI: 10.3390/buildings15101609

Doctoral Research (in progress)

Civil Engineering — Construction Management · Kasetsart University

# research interests

Computer vision for construction safetyDigital twins for infrastructureDeep learning & image super-resolutionAsset-lifecycle data interoperability
ORCID0009-0007-9632-9167

# theses

  • M.Eng

    ESPCN-YOLO: A High-Accuracy Framework for Personal Protective Equipment Detection Under Low-Light and Small Object Conditions

kowz@ekasithk:~$ education --list

Education

Ph.D.2026 – Present

Civil Engineering — Construction Management

Kasetsart University

  • ·Doctoral research in progress.
M.Eng.2024 – 2025

Civil Engineering — Construction Management

Kasetsart University

  • ·Thesis: ESPCN-YOLO — PPE detection under low-light & small-object conditions.
  • ·Published in MDPI Buildings (2025).
B.Eng.2019 – 2023

Civil Engineering

Kasetsart University

kowz@ekasithk:~$ ./contact --hire-me

Get in touch

Open to consulting & new opportunities. Whether it's a digital-twin build, a research collaboration, or consulting — let's talk.

Email:[email protected]Academic:[email protected]GitHub:https://github.com/EkasithKInstagram:https://www.instagram.com/ekasith.k/ORCID:https://orcid.org/0009-0007-9632-9167
📍 Bangkok, Thailand🕒 GMT+7
Download CV.pdf
QR code linking to ekasithk.com

scan to open

ekasithk.com