Hacker News .hnnew | past | comments | ask | show | jobs | submit | dawitworku's commentslogin

I built Gityap — a small tool that tries to answer a simple question:

Do you ship more than you talk?

Gityap compares a GitHub user with a Telegram channel and outputs a single score: a clean ship-to-talk ratio.

No dashboards. No analytics maze. Just signal.

The Idea

There’s a lot of noise in tech.

Threads. Opinions. Announcements. “Building in public.”

But how much actual code is being shipped?

Gityap connects:

GitHub commits

Telegram posts

And produces one clear output:

Signal score = commits vs. posts

You immediately see creator behavior patterns.

What It Does

Compare a GitHub handle + Telegram channel

Output one score

Show ship ratio

Rank top GitHub “shippers”

Rank top Telegram “yappers”

Track most active channels by subscribers

Optional:

Connect GitHub OAuth to include private commits (for your own account)

Public-only works without login.

Example Data

Top GitHub Shippers (by commits)

@levlam — 259,448 commits

@Dawaman43 — 6,028 commits

@frectonz — 4,866 commits

@sifenfisaha — 1,737 commits

@natanimn — 1,570 commits

Top Telegram Yappers (by posts)

@gugutlogs — 4,757 posts

@cyrilogban — 4,220 posts

@thefrectonz — 3,855 posts

@BurhanOps — 2,602 posts

@DoughNutDrops — 2,376 posts

It’s interesting to see overlaps — some builders talk a lot, some barely talk, some do both.


I’ve been using the Internet Archive a lot recently and found the default UI pretty slow and painful for multi-file items and bulk downloads.

This tool provides a much cleaner experience:

Server-side search & pagination

Better previews for long playlists/episodes

Reliable browser downloads via a proxy

One-click ZIP downloads

Keyboard-friendly media player

Open source and community-built. Source: https://github.com/Dawaman43/downloadstuffs

Not affiliated with Internet Archive.


Stop writing boilerplate! SpeedyEDA gives you instant insights, visualizations, and summaries from any dataset in one command.

Features include:

Automatic statistics: mean, median, std, correlations

Missing value analysis with actionable suggestions

Auto visualizations: histograms, boxplots, heatmaps

Colorful, emoji-rich terminal output with Fun Mode

Presets for ecommerce, surveys, and finance

Plugin system to extend with custom analyses

Interactive mode and batch processing for multiple datasets

Works from the CLI or Python API:

pip install speedyeda fasteda data.csv --fun --plots

Perfect for data scientists, analysts, and Python developers who want fast, beautiful exploratory data analysis without setup.


UltraDL Pro is a cross-platform downloader with both a Linux TUI and a Python CLI that works on Linux, macOS, and Windows. It supports multiple platforms including YouTube, TikTok, Twitter, Twitch, and Spotify.

Key Features:

Multi-platform video support

Spotify tracks, albums, and playlists download

Terminal-based YouTube search

Auto-organizer for downloads

Browser cookie integration to bypass bot detection

Multi-threaded fast downloads using aria2c

Interactive UI powered by gum

Installation:

python -m pip install -U ultradl-pro ultradl-pro download <url>

Optional: For better isolation, install via pipx. Some features require ffmpeg.

GitHub: https://github.com/Dawaman43/ultradl-pro


Former Netflix engineer ThePrimeagen has become one of the most influential developer educators online. This profile breaks down his background, teaching philosophy, Vim obsession, and why so many developers resonate with his focus on fundamentals, performance, and intentional software design.


Author here

xsql started as a small internal tool to avoid rewriting SQL schemas by hand when switching between SQLite, PostgreSQL, and MySQL for local development and testing.

The main idea is to parse CREATE TABLE statements into a small intermediate representation (IR) instead of doing text-based rewrites. That makes the conversions easier to test and extend, and it’s why I’m experimenting with a richer IR v2 (including constraints and JSON output) rather than adding lots of ad-hoc rules.

The scope is intentionally limited for now (schema DDL only, no data migration). I’d really appreciate feedback on the IR approach, portability edge cases, or what features matter most in real-world schema migrations.


Hi HN,

I'm the creator of Memarya. I started this project because I believe that access to education should be open and free for every student. My goal is to build a full-featured, open-source e-learning platform that communities can host and adapt themselves.

It's still in the early stages, but I've just opened up the source code and would love to build this out in the open. I'm looking for feedback, ideas, and collaborators who are passionate about open education.

The project is built with Next.js, Drizzle ORM, Auth.js, and shadcn/ui. I'd be grateful for any feedback on the code or architecture.

If you're a developer, educator, or designer who finds this interesting, please check out the repository. I'm setting up a roadmap in the GitHub Issues, and I welcome any contributions, from bug reports to new features.

Thanks for checking it out!


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: