HNHacker News
TopNewBestAskShowJobs

flavius37

33 karma · joined May 20, 2017

submissionscomments
flavius37··on Ask HN: What's your current AI agentic setup?
Here is a summary of the comments from this thread. Updates will follow to include latest comments too.

https://techaddo.com/posts/ai-agentic-setups

flavius37··on Awesome HN: The sites that mine its comments for picks
Here is the first site of this kind: https://web.archive.org/web/20110305143421/http://www.hacker...

There is Section 5 that is mentioning the flaws:

- A mention isn't a recommendation.

- Counts track popularity, not quality

- They inherit HN's taste.

- Picks go stale.

- The sites don't last

- Matching misfires. Unlinked names are ambiguous.

flavius37··on Show HN: HackerTV – HN's most-recommended Youtube videos as 24/7 TV channels
Here comes the hacker of hackers of hackers... : https://news.ycombinator.com/item?id=50362372
flavius37··on Awesome HN: The sites that mine its comments for picks
HackerTube set off a dozen spin-offs mining Hacker News comments for books, papers and more. The idea dates to 2011; its flaws were spotted by 2013.
flavius37··on Ask HN: What's your choice for the classic web app stack?
Depending on the app type:

* If the app it's a CRM like (with lots of DB entities) I will go with Python Django, MySQL, vue.js

* For Browser apps or desktop apps that has few endpoints I will go with Golang (https://go-app.dev/) for backend and frontend.

* For a mobile app: React, the backend depends on performances, Golang or Python Django.

* IoT app (may need to process sensors data, analyze, and visualize): Golang, ClickHouse, vue.js

If it's a very simple app (e.g. a dashboard) I will be using: https://go-app.dev/ with SQLite.

flavius37··on Letting neural networks be weird
Here is explain a paper that uses AI to gind bugs in games: https://youtu.be/wm8tK91k37U