DEV Community

Japan Open Data Lab profile picture

Japan Open Data Lab

Fighting Kaggle with Gemini (Store Sales #3/612, Fraud #142/1,519). 150-item sourced knowledge vault. 'What prefecture are you?' stats quiz: jpopendata.jp. Fully audited data.

Location Tokyo, Japan Joined Joined on  Personal website https://rapidapi.com/organization/japan-opendata-lab github website twitter website

Education

Computer Science & Quantitative Economics

Pronouns

they/them

Work

Data Engineering & Research Team at Japan Open Data Lab

What Is Your Prefectural Blend? A Quiz Computed from Real Government Statistics
japanopendatalab profile

What Is Your Prefectural Blend? A Quiz Computed from Real Government Statistics

#showdev #datascience #japan #webdev
1
Comments
3 min read

Want to connect with Japan Open Data Lab?

Create an account to connect with Japan Open Data Lab. You can also sign in below to proceed if you already have an account.

Create Account
Already have an account? Sign in
I Built a 150-Paper Knowledge Vault to Fight Kaggle with Gemini Alone
japanopendatalab profile

I Built a 150-Paper Knowledge Vault to Fight Kaggle with Gemini Alone

#kaggle #machinelearning #gemini #python
Comments
4 min read
Thought Experiment: If 100GB of structured/3D data could be losslessly compressed to 1GB, what would you build?
japanopendatalab profile

Thought Experiment: If 100GB of structured/3D data could be losslessly compressed to 1GB, what would you build?

#discuss #architecture #dataengineering #database
Comments
3 min read
Building an Automated Python ETL Pipeline for Japan's 1,741 Municipalities Open Data (SSDSE-A)
japanopendatalab profile

Building an Automated Python ETL Pipeline for Japan's 1,741 Municipalities Open Data (SSDSE-A)

#automation #datascience #python #tutorial
1
Comments
6 min read
What I Learned Building Multiple MCP Servers for Local LLMs (Qwen3-8B): Tool Granularity, Latency Traps, and Dogfooding
japanopendatalab profile

What I Learned Building Multiple MCP Servers for Local LLMs (Qwen3-8B): Tool Granularity, Latency Traps, and Dogfooding

#mcp #llm #python #duckdb
2
Comments 2
5 min read
MCP Explained in 5 Layers: How I Connected 10.16M Records to AI (and Why It's a Blue Ocean)
japanopendatalab profile

MCP Explained in 5 Layers: How I Connected 10.16M Records to AI (and Why It's a Blue Ocean)

#mcp #claude #ai #opensource
1
Comments 4
4 min read
Analyzing 40 Years of Japanese Public Open Data in 3 Minutes with DuckDB (No API Keys, In-Memory Parquet)
japanopendatalab profile

Analyzing 40 Years of Japanese Public Open Data in 3 Minutes with DuckDB (No API Keys, In-Memory Parquet)

#python #datascience #duckdb #ai
Comments
3 min read
loading...