👨🏻‍🦳 The Winter Maintenance Optimization Tool

https://interactive-or.net/snowplowpro

... is a web-based decision support system that automates these processes using arc-routing algorithms—advanced optimization methods designed for servicing specific segments of a road network under time and resource constraints. The tool generates efficient plow and salt routes, aligned with operational goals and regulatory requirements.

https://www.tradingview.com/watchlists/123254792/

Basically, a list of potential employers worldwide :)

https://airtable.com/embed/apphVYa9Taubt6wPl/shryZqOJy2O8CMaLQ — very detailed list of Nordic Mobility Startups (650+) (AirTable table) monitored and documented by https://betamobility.com/resources

Geoff Boeing’s new paper: «Topological Graph Simplification Solutions to the Street Intersection Miscount Problem»

https://onlinelibrary.wiley.com/doi/10.1111/tgis.70037

Geospatial Mechanistic Interpretability of Large Language Models

arXiv paper: https://arxiv.org/abs/2505.03368

GitHub code: https://github.com/sdesabbata/geospatial-mechanistic-interpretability


Hf: https://huggingface.co/papers/2505.03368

that 10x dev guy
pure gem for ds crew

https://github.com/koaning

he also did this..
I am shocked how good this is

https://calmcode.io

basically a list of tiny things that will make your pythonic life a bit more comfy

😱

🐟Fresh!🎣

TLDR they literally mapped all the buildings on the globe
——

GlobalBuildingAtlas LoD1 by TUM

gh: https://github.com/zhu-xlab/GlobalBuildingAtlas

explore!: https://tubvsig-so2sat-vm1.srv.mwn.de

read: https://arxiv.org/pdf/2506.04106

The full dataset will be completely open soon here: https://mediatum.ub.tum.de/1782307 (coming soon)
——————————-

Mostly precise, but not ideal. Especially in high-rise areas

https://medievalmurdermap.co.uk

hehe classic

Сравнение данных OSM за разные периоды: воспроизводимый код и небольшое описание того, как он был создан.

Модуль для QGIS для сравнения между собой различных слоев или их наборов - QMapCompare.
Описание модуля и документация

https://github.com/marin-community/marin?tab=readme-ov-file

Marin is an open-source framework for the research and development of foundation models.

A key feature of Marin is reproducibility: every step, from raw data to the final model are recorded, not just the end result. This includes failed experiments, so the entire research process is transparent.

Marin's primary use case is training language model like Llama, DeepSeek, Qwen, etc. Notably, this includes data curation, transformation, filtering, tokenization, training, and evaluation.

We used Marin to train the first open-source 8B parameter model to outperform Llama 3.1 8B. You can see the training script or read the retrospective.

https://github.com/HydroPML/FloodCastBench

https://www.nature.com/articles/s41597-025-04725-2

a large-scale, ML-ready dataset for dynamic flood modeling and forecasting

hot topics in urban morphology rn
https://www.isuf2025.org/wp-content/uploads/2025/06/ISUF2025_PROGRAM250619.pdf

URBAN MORPHOLOGY IN THE AGE
OF ARTIFICIAL INTELLIGENCE
XXXII International Seminar On Urban Form
16th-21st June 2025