frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Talos: Hardware accelerator for deep convolutional neural networks

https://talos.wtf/
41•llamatheollama•2h ago

Comments

arjvik•1h ago
Not to take away from this cool project, but its design decisions are incredibly impractical.
refulgentis•1h ago
I honestly can’t tell if it’s a cool project or just a md file someone with 0 experience had an LLM output.
noosphr•1h ago
> It isn't just a reimplementation of existing software logic in hardware; it is a rethinking of how deep learning inference should work at the circuit level. [...] By implementing the entire inference pipeline in SystemVerilog, we achieve deterministic, cycle-accurate control over every calculation. [...] But don’t let the two-week timeline fool you. Those were two weeks full of 18-hour days, fueled by caffeine and sheer stubbornness.

I'm having a hard time figuring out if this is satire or not.

jcgrillo•33m ago
From personal experience caffeine is not enough for 2wk of 18hr days.. you need some pervitin type shit
tadfisher•1h ago
My advice: write your own English prose, and try not to let "LLM-speak" leak into your documentation when using them to edit. Ironically, LLMs just plain suck at writing English, like they're incredibly overfit on marketing copy and press releases. I hope someone is working on this, or at least cares about the problem, because that would make this brave new world palatable for reading.
vidarh•1h ago
They really don't, if you actually bother prompting them. Give them a voice sample, and tell them to match the tone, and you already get something 10x better. Have them revise with a list of common writing problems - not just common LLM patterns, but guidelines for writing better - and you get rid of more.

Properly prompted, an LLM writes far better than most people.

jdcasale•49m ago
Without weighing in on whether this is true, I'll point out that LLMs could both be better writers than most people and also be bad writers.

Writing is a difficult skill that many (most?) educational systems do not effectively teach. Most people are terrible writers.

tadfisher•46m ago
That is precisely the problem. When writing technical documentation, such as the landing page for an FPGA inference engine, a model should not need to be prompted to use proper voice and to avoid marketing language. There should be enough context in the text of the prompt itself.
bonoboTP•18m ago
I don't think any of this indicates a fundamental property of the tech itself. AI companies post-train their models to sound like what people like to read better. There's a reason that engagement farmers have converged on the tone that these LLMs imitate, namely its something that people prefer. Maybe not you, but it's the same thing that gives us YouTube face on thumbnails etc.

It takes some prompting to nudge the model out of that default voice because post training reinforced it. They will likely shift it once these AI-isms are known and recognized widely. I'd assume the nextgem models under training now will get negative feedback from the human evaluators for talking too AI-like and then there will be new AI smells to calibrate to.

Wowfunhappy•8m ago
I don't understand what "should" means in this context.

Personally, I still can't quite believe that any of this is possible. A computer should be unable to write any prose at all. If it can, but it defaults to a certain tone of voice and needs to be prompted a bit to do something else... okay? I guess that's how it is.

refulgentis•1h ago
This is horrible LLM slop, my god.

Winced my way through “Convolutions are in CNNs (it’s literally in the name, Convolutional Neural Network)”, then had to stop.

It’s honestly offensive to me. It doesn’t even make sense on its own terms. For some reason we fly from LLM inferencing to toy MSINT to convolutions with __0__ transition or sense of structure.

fc417fc802•44m ago
Aside from the verbose AI slop it's an interesting hobby project for exploring FPGAs. But it doesn't do anything you can't do on CPU by using a model that's small enough to fit in cache. In terms of practical use you'd be better off implementing a minuscule model using vector intrinsics in your favorite systems language.
roughly•47m ago
If the author and/or anyone else hasn't seen Sidero's Talos Linux distro, it's my current favorite way to spin up a bare metal Kubernetes cluster:

https://www.talos.dev/

neoCrimeLabs•9m ago
Agreed.

Also, in my experience, a great way to run K8s in IAAS while minimizing vendor lock-in.

zmmmmm•39m ago
> Talos is a custom FPGA-based hardware accelerator built from the ground up to execute Convolutional Neural Networks with extreme efficiency

Makes it sound like it's new hardware. This is just (I'm inferring) software to program an off the shelf FPGA to do convolutions. Very minimal ones by the look of it (MNIST etc).

MarcelOlsz•15m ago
Love those animations/diagrams. How were they made?

MacBook Pro with M5 Pro and M5 Max

https://www.apple.com/newsroom/2026/03/apple-introduces-macbook-pro-with-all-new-m5-pro-and-m5-max/
652•scrlk•11h ago•642 comments

GPT‑5.3 Instant

https://openai.com/index/gpt-5-3-instant/
285•meetpateltech•7h ago•208 comments

Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

https://www.tomshardware.com/pc-components/cpus/intels-make-or-break-18a-process-node-debuts-for-...
237•vanburen•6h ago•189 comments

Lenovo's New ThinkPads Score 10/10 for Repairability

https://www.ifixit.com/News/115827/new-thinkpads-score-perfect-10-repairability
115•wrxd•1h ago•28 comments

Textadept

https://orbitalquark.github.io/textadept/
61•giancarlostoro•2d ago•11 comments

Claude's Cycles [pdf]

https://www-cs-faculty.stanford.edu/~knuth/papers/claude-cycles.pdf
448•fs123•14h ago•207 comments

Voxile: A ray-traced game made in its own engine and programming language

https://elbowgreasegames.substack.com/p/voxray-games-pushes-major-update
89•spacemarine1•4h ago•16 comments

An Interactive Intro to CRDTs (2023)

https://jakelazaroff.com/words/an-interactive-intro-to-crdts/
89•evakhoury•6h ago•15 comments

The Xkcd thing, now interactive

https://editor.p5js.org/isohedral/full/vJa5RiZWs
1135•memalign•14h ago•152 comments

What’s in a name? (2014)

https://sailsandcommas.com/2014/02/03/whats-in-a-name/
9•Curiositry•2d ago•1 comments

When AI writes the software, who verifies it?

https://leodemoura.github.io/blog/2026/02/28/when-ai-writes-the-worlds-software.html
134•todsacerdoti•8h ago•128 comments

Time, Space, and Life as We Know It (2017)

https://raganwald.com/2017/01/12/time-space-life-as-we-know-it.html
5•vismit2000•3d ago•0 comments

We've freed Cookie's Bustle from copyright hell

https://gamehistory.org/cookies-bustle/
87•sb057•5h ago•10 comments

Don't become an engineering manager

https://newsletter.manager.dev/p/dont-become-an-engineering-manager
302•flail•11h ago•217 comments

Physics Girl: Super-Kamiokande – Imaging the sun by detecting neutrinos [video]

https://www.youtube.com/watch?v=B3m3AMRlYfc
419•pcdavid•10h ago•66 comments

Launch HN: Cekura (YC F24) – Testing and monitoring for voice and chat AI agents

70•atarus•10h ago•19 comments

TorchLean: Formalizing Neural Networks in Lean

https://leandojo.org/torchlean.html
73•matt_d•2d ago•9 comments

Don't make me talk to your chatbot

https://raymyers.org/post/dont-make-me-talk-to-your-chatbot/
208•pkilgore•3h ago•162 comments

Talos: Hardware accelerator for deep convolutional neural networks

https://talos.wtf/
41•llamatheollama•2h ago•16 comments

Possible US Government iPhone-Hacking Toolkit in foreign spy and criminal hands

https://www.wired.com/story/coruna-iphone-hacking-toolkit-us-government/
182•alwillis•5h ago•55 comments

Disable Your SSH access accidentally with scp

https://sny.sh/hypha/blog/scp
98•zdw•3d ago•45 comments

I'm reluctant to verify my identity or age for any online services

https://neilzone.co.uk/2026/03/im-struggling-to-think-of-any-online-services-for-which-id-be-will...
877•speckx•11h ago•537 comments

MacBook Air with M5

https://www.apple.com/newsroom/2026/03/apple-introduces-the-new-macbook-air-with-m5/
370•Garbage•11h ago•417 comments

Wait what, you can't use a *code* editor when you're under 18 now?

https://mastodon.online/@marekfort/116164253291515471
8•pabs3•31m ago•0 comments

I'm losing the SEO battle for my own open source project

https://twitter.com/Gavriel_Cohen/status/2028821432759717930
445•devinitely•11h ago•221 comments

TV's TV (1987) & TV Games Encyclopedia (1988)

https://blog.gingerbeardman.com/2026/03/01/tvs-tv-1987-and-tv-games-encyclopedia-1988/
10•msephton•2d ago•0 comments

Meta’s AI smart glasses and data privacy concerns

https://www.svd.se/a/K8nrV4/metas-ai-smart-glasses-and-data-privacy-concerns-workers-say-we-see-e...
1369•sandbach•1d ago•770 comments

The Two Kinds of Error

https://evanhahn.com/the-two-kinds-of-error/
33•zdw•2d ago•19 comments

Apple Studio Display and Studio Display XDR

https://www.apple.com/newsroom/2026/03/apple-unveils-new-studio-display-and-all-new-studio-displa...
214•victorbjorklund•11h ago•249 comments

The beauty and terror of modding Windows

https://windowsread.me/p/windhawk-explained
112•wild_pointer•14h ago•88 comments