Hacker News Reader

eamagon Google uses AI to reduce stop-and-go traffic on your route

Maybe here? https://blog.google/outreach-initiatives/sustainability/goog...

> We build an AI-based model of each intersection, including its structure, traffic patterns (such as patterns of starting and stopping), light scheduling, and how traffic and the light schedules interact, and then we also build a model of the interaction between traffic lights. Based on this model, we develop AI-based optimizations and then provide recommendations to city engineers via the Green Light interface. As an example, we might identify an opportunity to coordinate between intersections that are not yet synced and provide a recommendation around the timing of the traffic lights so that traffic flows more effectively along a stretch of road.

eamagon Ask HN: What have you worked on this year?

The last one I built is an interactive map of all institutions that published at the International Conference on Machine Learning this year, https://arxiv-cat.eamag.me/

eamagon Trunk: Build, bundle and ship your Rust WASM application to the web

Are there any examples?

eamagon Microsoft laid off a DEI team, DEI is 'no longer business critical'

I couldn't submit it because the original title is too long

eamagon Ask HN: How do you find ideas to work on?

I have two approaches:

1. Random ideas that come from reading, browsing, coding - like https://eamag.me/2024/HackerNews-Prediction-Evaluator#why

2. Have some big goal in mind and think about missing things that have to be built to achieve it - like https://eamag.me/2024/Automated-Paper-Classification#motivat...

eamagon Cradle: Empowering Foundation Agents Towards General Computer Control

I'm looking through the code, does it mean that the authors wrote all basic skills themselves and let LLM choose from them? So this approach can't be generalised, can it?

https://github.com/BAAI-Agents/Cradle/pull/44/files#diff-3f3...

eamagon The most cited NLP papers each year

How did you identify what is NLP?

eamagon Evaluating a Decade of Hacker News Predictions: An Open-Source Approach

You can go to the table [1] page and filter by username, then you can see what prediction LLM is evaluating and why. I can agree with you on hallucination part, but for the easy predictions it's correct most of the time (like Microsoft downfall)

[1]https://hn-predictions.eamag.me/table

eamagon ClickHouse, Inc.

There is a video (in russian) [1] with an idea "if someone claims something works faster than clickhouse - it means I haven't optimised this specific query yet" [1] https://www.youtube.com/watch?v=MJJfWoWJq0o