PT
Praruj Thapa
AI/ML Engineer & Data Scientist
/
About

I'm an ML engineer and data scientist at Fukushima Galilei (フクシマガリレイ株式会社). I build predictive maintenance for refrigeration equipment: models that warn us about a failure a few days before it happens. I trained as an electrical and mechanical engineer, so I start from how the machine works and add the machine learning after. On the side I build projects in search, wearables and simulation, and I write up what went wrong as well as what worked. I work in English and Japanese.

Experience
ML & AI Engineer | Data Scientist Jun 2025 — Present

Fukushima Galilei (フクシマガリレイ株式会社)

I build predictive maintenance for commercial refrigeration and air-conditioning equipment. The pipeline flags failures 3–4 days ahead and is used every day. The features come from refrigeration-cycle physics (superheat, subcooling, pressures, compressor duty cycle). I separate defrost and door openings from real faults so false alarms stay low. A local LLM explains each forecast in plain language, and colleagues use it through FastAPI and Streamlit tools.

Production Engineer (Intern) Sep 2023

Fukui Megane (EssilorLuxottica) · Fukui

Podcast Host 2020 — 2022

Guffadi Cast · Freelance

Founder, Rotaract Club of Dakshinkali 2021 — Present

District 3292, Zone 11

Projects

Personal projects on public data, each with its code on GitHub. Every card says what it does, the result and the tools; “Case study” has the full story.

Featured
Research: behaviour & language
Health & wearables
More projects
Forecasting & simulation
Nepal Premier League 2026: 20,000 Simulated Seasons

The cricket sibling of the World Cup forecast: two models, one from team results and one ball by ball, play out the rest of the T20 season with the real net-run-rate tie-break and playoff bracket.

35%Lumbini Lions' title chance, the favourite as of 9 November
0.687Backtest log loss on 44 games (coin flip = 0.693; Elo alone did no better)
Built with Python · NumPy · Monte Carlo simulation · Elo ratings · Cricsheet ball-by-ball data · pytest
IoT & computer vision
Smart Showcase Management System
3rd-Year Project

Raspberry Pi + OpenCV shelf monitoring with LINE Notify alerts. Led directly to my current role. Fukushima Galilei, 2023–24.

Built with Raspberry Pi · OpenCV · Python · LINE Notify
Computer vision
Vision Transformer Projects

Two transformer-based CV projects: DETR hardhat detection with bipartite matching (loss 1.35), and ViT + knowledge distillation on Oxford Pets (11.6% → 37.6% accuracy). Personal, 2026.

Built with PyTorch · DETR · Vision Transformer · knowledge distillation
Skills
Languages & Tools
Python, SQL (MySQL), Git, GitHub Actions, pytest
ML & Data
PyTorch, scikit-learn, Hugging Face Transformers, pandas, NumPy, SciPy (signal processing), time-series modelling, Monte Carlo simulation
Search & Recommendation
BM25, multilingual embeddings (E5), hybrid retrieval, cross-encoder reranking, Japanese tokenization (MeCab), recommenders (EASE, ALS)
AI & Backend
RAG, local LLMs, FastAPI, Streamlit
Behavioral Science
Computational models of choice (reinforcement learning: PVL-Delta, VPP, ORL), model fitting and parameter recovery, statistics (nonparametric, permutation, bootstrap, multiple-comparison correction), LLM agents with chain-of-thought
Computer Vision & Hardware
Vision Transformers, DETR, OpenCV, Arduino, Raspberry Pi
Cloud
AWS (Certified AI Practitioner, Certified Cloud Practitioner)
Certifications
AWS Certified Cloud Practitioner
AWS Training & Certification · Oct 2026
Verify ↗
AWS Certified AI Practitioner
AWS Training & Certification · Sep 2026
Verify ↗
AWS Cloud Quest: Generative AI Practitioner
AWS Training & Certification · Aug 2026
Verify ↗
Education
Kyoto University of Advanced Science 2021 — 2025

B.E. Electrical & Mechanical Systems

Writing

I write up my research as case studies (causal ML, psychometrics, wearables) alongside essays on learning and language — read them here ↗

Contact

Open to work · Open to remote collaboration