frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Ask HN: How do you get into systems programming

13•otherayden•1y ago
Hi all!

I'm looking for recommendations on where to start with learning systems programming. Ideally, I'd like to be able to get to a point where I can make a living doing it, but currently I just want to do fun stuff to build up curiosity around it.

Here's all of the "low-level" stuff that I know so far / imagine being useful. I... - Have enough of an understanding of networking to write a toy HTTP server on top of TCP - Know enough C to write some basic terminal tools + window applications if needed (on Linux) - Love terminal tools like neovim + several core utils - Have dabbled with Arduino/ESP32 & communicating via USB over the serial port with a host pc - Am pretty decent with Python, and have been using it for like 10 years

Some things that I've been curious about in the past - Converting parts of python libraries from pure python to C/C++ bindings for better performance - Writing a terminal based file manager to work with Google Chrome - Actually contributing to chromium (my laptop is a potato though so all of my builds fail)

About me: I'm in my junior year of uni studying CS, and I've been able to make money doing web dev for the past 2 years of my degree. For many reasons including curiosity and the fact that AI makes me feel replaceable doing many frontend + backend tasks, though I'm very curious about getting into lower level programming.

Any help would be greatly appreciated!

Comments

abhisek•1y ago
IMHO there is neither baseline nor “enough” when it comes to learning any programming language for any reasonably complex domain.

As you already know, C/C++ helps with low level software layers that interface with or manage hardware resources. In my experience, Go and Rust are also pretty much used as systems programming languages. For example, I use Go and EBPF to instrument systems calls on Linux kernel.

For me, most of my learning came from solving problems and building for specific use-cases. I think getting into builder mode and creating some cool will definitely accelerate your learning.

sargstuff•1y ago
On software side, building an OS (distribution) from scratch provides a step above bare metal programming[0].

Provides familiarity with different types of things a kernel does via programs/scripts that make use of kernel.

Actually writing binary code for kernel bit can be done under qem[1][2]. aka don't need to buy actual hardware, can use 'software probes' to view what's going on, etc. Don't have to worry about 'crashing'/trashing box running on (just crash the qem software & loosing just what was done in qem session, if didn't save as 'export/save to external location outside of qem session')

"Reading OpenBSD source code daily (blog.tintagel.pl)" from [hn: 3] automated way to review code.

-----

[0] : https://www.linuxfromscratch.org/

[1] : qem for kernel developers - https://www.youtube.com/watch?v=LyWlpuntdU4

[2] : https://www.collabora.com/news-and-blog/blog/2017/01/16/sett...

[hn:3] : https://news.ycombinator.com/item?id=14521386

a_tartaruga•1y ago
It sounds like you're doing the normal sort of things that systems people do to get started. The fact that you have lots of ideas to jump off of is very good. In general just follow all of your ideas down as far as you can to the base systems. Write the TCP implementation for your HTTP server and run it over the internet for example. You've only gone too far when you start worrying about noise and debugging looks like randomly grounding metal things.
theophilec•1y ago
Oxide and Friends has an episode on the topic [1], I found interesting.

[1] : https://oxide-and-friends.transistor.fm/episodes/paths-into-...

noone_youknow•1y ago
Sounds like you’re doing some interesting stuff and have a good, varied skill base to build on.

My advice would be to jump in and start working on kernel level stuff, or writing your own - IMO there’s no finer way to really “get” the low level concepts and the understanding you’ll build will really help with any other system-level stuff you do.

Not to plug, but if you were interested in getting involved in an existing project, my own toy kernel project[0] is at a point where there’s still lots of fun stuff left to do (both design- and implementation-wise) but a lot of the basic “project plumbing” and one-time machine setup stuff that people often get stuck on is already done, and I’d be glad to have the opportunity to share knowledge.

[0] : https://github.com/roscopeco/anos

Show HN: Encarta 95 Web

https://benwinding.github.io/encarta_95_web/#/article/Art
1•gitgud•28s ago•0 comments

A supercomputer is about to launch to the moon aboard a rover named FLIP

https://www.space.com/space-exploration/a-supercomputer-is-about-to-launch-to-the-moon-for-the-1s...
1•pseudolus•2m ago•0 comments

Why moving sprites are slow

https://debuff.jp/blog/why-moving-sprites-are-slow
1•gregjw•2m ago•0 comments

We used RCE on 3M PCs to prevent a botnet

https://labs.eye.security/operation-ixim-preventing-a-botnet-of-millions-of-popcorn-time-users/
1•the1bernard•3m ago•0 comments

Desert Bus 2026

https://desertbus.org/2026
1•Tomte•3m ago•0 comments

Samsung TV on-device AI listening, using soundbar's mic if the TV has none

https://www.reddit.com/r/singularity/comments/1xduimx/hacker_reverse_engineers_his_samsung_tv_and...
2•saulapremium•5m ago•0 comments

The animals you can't outrun – and the ones you can

https://www.bbc.com/future/article/20261021-what-animals-could-you-beat-in-a-race
1•cisc•8m ago•0 comments

Few Americans Say They Are Asking AI Chatbots for 2026 Election Information

https://www.pewresearch.org/short-reads/2026/10/23/few-americans-say-they-are-asking-ai-chatbots-...
1•giuliomagnifico•8m ago•0 comments

Damage Control

https://fireborn.mataroa.blog/blog/damage-control/
1•u1hcw9nx•9m ago•0 comments

Rapidly scaling online storage to serve over 1B ChatGPT users (Part II)

https://openai.com/index/scaling-storage-one-billion-users-part-two/
1•frostbyte7•10m ago•0 comments

Apple Silicon Inference Engine: Kapitsa

https://github.com/nace-ai/kapitsa
1•fearandgreed•18m ago•0 comments

Iran has detailed plans to hit Gulf undersea cables if war resumes

https://www.reuters.com/world/middle-east/iran-has-detailed-plans-hit-gulf-undersea-cables-if-war...
2•geox•24m ago•0 comments

Attention people who think UB is fine

https://blog.habets.se/2026/10/Attention-people-who-think-UB-is-fine.html
1•praptak•31m ago•0 comments

Don't Embarrass Me, Claude

https://www.dev-log.me/dont-embarrass-me-claude/
2•wazHFsRy•35m ago•2 comments

Claude Duet Safety, Privacy, and Security

https://claude.com/resources/articles/claude-duet-safety-privacy-and-security
1•pretext•36m ago•1 comments

Spritz: Stream Local Videos via AirPlay, Cast and DLNA on Apple Silicon

https://github.com/Oolab-labs/spritz
1•kls0e•39m ago•0 comments

Retroviral insertions contributed to divergence of human and chimpanzee brains

https://www.science.org/doi/10.1126/sciadv.aeg5791
1•jmillikin•40m ago•0 comments

All existing implementations of softmax and cross-entropy have a subtle bug

https://twitter.com/jon_barron/status/2113707875290042665
1•pretext•42m ago•0 comments

Pair Prompt and Agentic Interview

https://liveshortly.com
1•malviyamukul•52m ago•0 comments

Show HN: Monkeytype for Developers

https://osdtype-w5vr-six.vercel.app/play
1•itsarnavsh•52m ago•1 comments

Show HN: Whyline – Git native decision history for coding agents

https://github.com/anishmoncivarghese/whyline
1•anishvarghese•52m ago•0 comments

Two companies raised three in every ten venture dollars

https://datahyena.com/research/two-companies-raised-three-in-ten-venture-dollars/
2•ghostfoxgod•53m ago•0 comments

Red Alert 2 Yuri's Revenge decompilation – rebuilding the lost source code

https://www.reddit.com/r/redalert2/comments/1xaep2v/ra2yr_decompilation_rebuilding_the_lost_source/
1•nomilk•55m ago•0 comments

Show HN: RetroBrag – remix Snake, Pong and Breakout by describing the change

https://www.retrobrag.com
1•likaiwei•58m ago•0 comments

That AI Temptation

https://www.tbray.org/ongoing/When/202x/2026/10/19/The-AI-Temptation
2•robin_reala•1h ago•0 comments

The attack chain behind the v1.107.0 LiteLLM release

https://docs.litellm.ai/blog/security-incident-oct-21-2026-attack-chain
2•feross•1h ago•0 comments

The Acceleration Paradox

https://medium.com/@smoking2man/the-acceleration-paradox-why-local-ai-driven-productivity-gains-f...
2•Aldus551•1h ago•0 comments

Abstract Code Art Animation – Evolution from Ones and Zeros to Neural Networks [video]

https://www.youtube.com/watch?v=-dTvmVUWY0o
1•jussehoo•1h ago•0 comments

California has more revenue than expected thanks to AI. Should it save it?

https://calmatters.org/politics/2026/10/california-proposition-2-explained/
2•MilnerRoute•1h ago•0 comments

The Fall of Scientific American

https://www.thenation.com/article/society/scientific-american-springer-nature-trump/
3•littlexsparkee•1h ago•0 comments