Personal Language Corpus
This workflow brings transcripts and exports into one small corpus, keeping passage IDs and source information attached to the analysis so a pattern can be checked against the words behind it.
Python
13 entries
2026
-
-
Lumen
Lumen connects an Obsidian vault to a local research backend, surfacing related notes and assembling reports whose citations are checked against the vault.
Source code -
Metlink Departures Display
A small departures sign for Wellington buses, built to put the next arrivals somewhere glanceable instead of repeating the same stop lookup.
Source code
2025
-
GrooveBox
GrooveBox is a small beat-sketching tool, using the computer keyboard to play samples and record patterns that can be changed while the sequencer runs.
Source code -
Bank-to-Sheets Sync
I made a small banking pipeline that keeps settled transactions in Google Sheets, leaving the budget calculations in the sheet and handling the recurring sync in Python.
Source code -
Cowrie Honeypot Analysis
I used Cowrie honeypot logs to examine the activity behind repeated login attempts, bringing event summaries and campaign-level observations into a pair of reports.
-
Pinterest Aesthetic Analysis
I used image embeddings to look at the themes inside a Pinterest board, grouping similar images and keeping snapshots that could be compared over time.
Source code -
TransitSync
TransitSync connected Google Calendar locations with Metlink route planning, keeping the travel problem close to the events that made the journey necessary.
Source code -
Sentient Environment Explorer
Sentient was an experiment in an agent learning its way through a changing 2D environment, with food and hazards giving its movement a consequence.
2024
-
Floor-plan Importer
This Blender add-on reads floor-plan data and creates the corresponding geometry, keeping the import step close to the modelling tools used to work on the result.
Source code -
Spotify Playlist Demographics
This experiment started with a Spotify playlist and followed its artists into biographical sources, using extracted metadata to build a view of the collection.
Source code
2023
-
ArticulateAI
ArticulateAI was one of my early voice experiments, recording a chosen amount of audio and passing it through transcription, a prompted response and optional speech output.
Source code
2021
-
Echo Stats Finder
I worked on Echo Stats Finder with Silveridge, connecting a Python desktop app to Echo VR and VR Master League so players could look up the people they were playing with.
Source code
