Praruj Thapa
ML & AI Engineer · Osaka, Japan
I build systems that notice patterns and predict what's next.
3–4 days early anomaly detection
90%+ mechanism reliability, tested
Daily used in production by analysts
Selected work

What I've built.

01
フクシマガリレイ · Industrial ML 2025
Time-Series Analysis Pipeline

Starts from raw, noisy sensor streams: cleaning, feature engineering, and exploratory analysis — heatmaps and figures — to identify which signals actually predict a coming fault, before any of it reaches a model. Forecasts machine state and flags anomalies 3–4 days before they happen. A locally-hosted LLM, fine-tuned on internal domain knowledge, turns the forecast into a plain-language read that data analysts can act on immediately. Used daily for fault analysis across multiple equipment types, and has held up reliably in production.

Python · Pandas · Feature Engineering · EDA · Time Series · Local LLM (fine-tuned) · Streamlit · MySQL
02
Personal Project · GitHub 2026
World Cup 2026 Prediction Engine

An Elo-based match model with Monte Carlo simulation — a real group stage, three-way win/draw/loss outcomes, and a live bracket that simulates forward from FIFA's actual knockout stage. Called the Final for Spain at 57.8%, validated against the real result (Brier score 0.178).

Python · Monte Carlo Simulation · Elo Ratings · pytest (18 tests)
View on GitHub ↗
03
フクシマガリレイ · Computer Vision 2023–24
Smart Showcase Management System

Real-time shelf monitoring with Raspberry Pi and OpenCV — auto-detects low stock and sends LINE Notify alerts. This project is what led directly to my current role.

Python · OpenCV · Raspberry Pi · LINE Notify API
Pre-Capstone Exhibition Award
Earlier engineering — Capstone, Shimadzu Corporation (2024–25)

An automated opening/closing mechanism for 15–50cl screw-type lab containers, built with SolidWorks CAD, Arduino, servo motors, IR sensors, and electromagnets. Tested for cycle time and reliability: under 3 seconds in 90%+ of trials, with an overall success rate above 90%.

Capstone Exhibition Award 2025
SolidWorks CAD · Arduino · Servo Motors · IR Sensors · Electromagnets
About

Nepal taught me patience. Japan taught me precision.

Before I wrote a line of Python, I spent two and a half years hosting a podcast — not because I wanted to be on air, but because I was obsessed with understanding people: why they feel what they feel, what they're really saying beneath the surface. That obsession didn't disappear when I switched to engineering. It just found a new medium.

“Data is just the world, speaking in numbers. I build systems that listen.”

At フクシマガリレイ株式会社, that's the discipline behind every pipeline I ship: observe first, understand what the data is actually saying, then build something a non-technical colleague can rely on without ever opening Python.

Currently at
フクシマガリレイ株式会社
ML & Software Developer · Osaka
Education
Kyoto University of Advanced Science
B.E. Electrical & Mechanical Systems · 2025
Languages
English · 日本語
Fluent · Fluent
Also
Rotaract Founder · Podcast Host
PyTorch · Scikit-learn · FastAPI · Streamlit · NumPy · Arduino · OpenCV
Experience

Where I've grown.

2025 — Present
ML & Software Developer
フクシマガリレイ株式会社 · Osaka, Japan

Building ML models, running time-series and data analysis, and developing the full software stack — FastAPI, Streamlit, MySQL — that delivers AI to the people who need it.

Sep 2023 — 1 mo
Production Engineer (Intern)
Fukui Megane (Essilor Luxottica) · Fukui, Japan

Production engineering internship focused on BOM management, routing optimization, and interplant communication — used data-driven problem solving to improve production data quality.

May 2020 — Sep 2022
Podcast Host
Guffadi Cast · Freelance

Hosted a Nepali podcast for 2.5 years. Built real skills in communication, storytelling, and understanding audiences — skills that turned out to be surprisingly relevant in engineering.

2021 — Present
Founder & Former Secretary
Rotaract Club of Dakshinkali · District 3292 (Nepal & Bhutan), Zone 11

During the 2020 lockdown, youth in the neighborhood had lost touch with each other. Co-founded a club to rebuild it: gathered 30 members from the municipality and started running sanitation, awareness, and fellowship programs for local youth — including a sanitation initiative at Champadevi Temple, one of several the club has run since.

Field notes

What I'm thinking about.

Current exploration
Can we build AI systems that know what they don't know?

Most ML models output a prediction with misplaced confidence. Reliable AI — systems that quantify their own uncertainty and defer appropriately — is one of the most underexplored problems in applied ML. I'm currently exploring conformal prediction and calibration as practical paths toward it.

Time-series intelligence
What physical systems reveal when you listen to them over time rather than sampling snapshots.
Local-first AI
Deploying capable models on-premises where data sovereignty and latency actually matter.
Human-centered automation
Systems that augment how skilled humans work, rather than pretending to replace their judgment.
I also write about literature and ideas at Tales of Shades ↗
Education

Foundations.

Engineering
Kyoto University of Advanced Science
B.E. Electrical & Mechanical Systems
2021 — 2025
Secondary
DAV College, Nepal
High School Certificate
2017 — 2019
Get in touch

Let's talk.

Open to conversations about ML systems, industrial AI, research collaborations, or just an interesting problem you haven't solved yet.

Based in Osaka · Open to work · Open to remote collaboration