Methods
8 articles
47 Scripts to 15: Cleaning a Research Codebase
Research projects accumulate code debt. An AI agent can map dependencies, identify dead code, and reorganize months of accumulated scripts, with…
Data Quality: From 49% to 12% Mobility Deserts
We found 49% of California census tracts were mobility deserts. After getting complete transit data, the corrected figure is 12%.
Robust API Collection: Pagination, Limits, Retries
Collecting location data from Google Places API at scale requires handling rate limits, pagination, and failure recovery.
Transit Travel Time Matrix in Python (Free, No API Keys)
Compute transit travel-time matrices in Python with r5py, using free GTFS and OpenStreetMap data. No API keys, fully worked example.
400 Labels to 94% Accuracy: Validating Grocery Store Data
Google Places returned thousands of 'grocery stores.' Many weren't. Here's how a classifier separates real grocery stores from gas stations and liquor stores.
From 7 Copy-Paste Cycles to 1 Command with AI Agents
The difference between chatbot-based coding and agent-based coding is categorical, not incremental. Here's what changes when AI can read your entire codebase.
From Methods Paragraph to Working Pipeline with AI
A well-written methodology section is almost executable code, and AI assistants have narrowed the gap between describing a procedure and implementing it.
One Context File: Teaching AI Your Research Project
Every new coding session with AI starts the same way: re-explaining the project. A single CLAUDE.md file loads automatically every session.