Writing

Notes

Recommenders, data pipelines and event cameras, worked through one problem at a time.

Histogram of pairwise cosine distances between event embeddings, peaking near 0.16, with the 0.09 cluster-match threshold marked to the left of the mass.
Recommenders 18 Aug 2026 · 24 min

How to build a recommender system in Postgres

A walk through the decisions behind a personalized feed, from representing somebody's taste to ranking for a group of four. Options, trade-offs and implementation for each, with Hango as the worked example.
Pipeline diagram: a dozen feeds and ninety thousand raw rows normalized onto one schema, resolved against what is stored, then merged or kept as new, leaving eleven thousand templates.
Data pipelines Next

How to deduplicate scraped listings with an LLM

Normalizing records from a dozen sources that disagree, resolving them against what is already stored, and deciding which near-duplicates are the same event. Hango's ingest as the worked example.

Two night-time frames of the same road side by side: the frame camera washed out by glare, the event camera still resolving the lane and reflector posts.
Perception Later

How to keep a vision model working after dark

Why a model trained on event-camera data barely notices when the lights go out, what that costs elsewhere, and when the swap is worth making. From an ACRA 2025 paper on the day-night domain gap.