frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Epic Games' new vibe-coded launcher is 46% faster than Steam

https://www.neowin.net/news/epic-wants-you-to-ditch-steam-for-its-new-faster-vibe-coded-games-launcher/
2•bundie•50m ago

Comments

biglyburrito•49m ago
And I'm sure it still sucks compared to Steam.
paimapi•44m ago
Would love for someone to pentest the launcher. Every company being this AI-pilled is really just asking for it when they brag about their vibe-coded products.
everdrive•40m ago
Epic is just setting up for the same rug-pull as every other tech company: get you entrenched, and then turn the screws. I don't care what they give away for free or what improvements they make.
tencentshill•32m ago
The steam client has always been a bit clunky, that's not the issue. They don't seem to be able to understand that.
nitwit005•21m ago
> The results reported that Epic's client is 46% faster than Valve's client during cold starts, clocking in at 5.47 seconds compared to Steam taking 10.07 seconds. When you restore both apps from system tray background states, the new client takes 0.36 seconds while Steam takes 0.42 seconds, according to the report.

> There are still missing features that the legacy client currently has, like comprehensive social panels and advanced account management options, with its developers "working toward feature parity."

Seems fairly likely it'll get slower as they add all the missing features. I'm not sure a couple of seconds startup time will be relevant to users.

Social Security benefits increase will hit 3.6% in 2027

https://rhodeislandcurrent.com/2026/10/14/repub/social-security-benefits-increase-will-hit-3-6-in...
1•chmaynard•28s ago•0 comments

Algorithmic Patterns on the Live Loom

https://www.matteringpress.org/books/homo-textor
1•bojackh•2m ago•0 comments

Sealed execution environment for GitHub Actions

https://github.com/nolabs-ai/runseal
1•decodebytes•2m ago•0 comments

LLM vs. Decision models like Jev and use-cases

https://grigio.org/llm-vs-decision-models-like-jev-and-use-cases/
1•grigio•3m ago•0 comments

Auditory processing strategies in humans and zebra finches

https://www.nature.com/articles/s42003-026-11035-9
1•bryanrasmussen•3m ago•1 comments

We should reject AI in academia

https://aeporreca.org/blog/why-we-should-reject-ai-in-academia/
1•speckx•4m ago•0 comments

Show HN: Checkobot – Detect Deepfakes and AI Images

https://checkobot.ai
2•krtk00•6m ago•0 comments

Show HN: OSS AI engineering telemetry without collecting prompts or code

https://github.com/ascendaone-com/ai-engineer-tools
1•emotf•7m ago•0 comments

Anthropic's Scion: A Four Year Project in Five Months

https://secondthoughts.ai/p/anthropics-scion-software-development
1•gk1•7m ago•0 comments

Show HN: Stateofpixel for UI Regression Testing

https://stateofpixel.com
1•vaibhavacharya•8m ago•0 comments

Show HN: Dub Anything – Dub videos in each actor's own voice

https://github.com/swax/dub-anything
2•swax•8m ago•0 comments

Jev Router: From Frontier Top Open Models

https://x.com/1littlecoder/article/2110390347977330961
1•amrrs•11m ago•0 comments

Casey Newton and John Gruber Slink Back to X

https://karlbode.com/casey-newton-and-john-gruber-slink-back-to-elon-musks-right-wing-propaganda-...
2•colinprince•14m ago•0 comments

Analyzing technical documents with Deep Review

https://research.google/blog/analyzing-technical-documents-with-deep-review/
2•adt2bt•14m ago•0 comments

Lockstep to pipeline in almost one line

https://andersmurphy.com/2026/10/12/lockstep-to-pipeline-in-one-line.html
1•speckx•16m ago•0 comments

Bank of America hired 1000 contractors to patch software flaws exposed by Mythos

https://www.cnbc.com/2026/10/14/anthropic-mythos-bank-of-america-hiring-cost.html
3•mirzap•17m ago•0 comments

Epic Games is bragging about their new store being vibe-coded

https://www.destructoid.com/epic-games-store-ai-vibe-coded/
3•ortusdux•18m ago•1 comments

The Coming Programming Language Renaissance

https://joshlf.com/posts/pl-renaissance/
1•alex_hirner•18m ago•0 comments

Reliable GPU Compute for Large Training and Inference Jobs

https://netflixtechblog.medium.com/reliable-gpu-compute-for-large-training-inference-jobs-d62b03a...
1•rutigs•19m ago•1 comments

Wealth Inequality in America Drawn to Scale

https://dreamagainusa.org/
2•laurex•19m ago•0 comments

Gemini 4 Argon

https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-argon/
3•ycosynot•19m ago•1 comments

How to Instrument a Multi-Agent System for Complete Causal Lineage

https://medium.com/@MirArshadTalpur/how-to-instrument-a-multi-agent-system-for-complete-causal-li...
1•Arshad-Talpur•20m ago•0 comments

AI agents overstate their results and remain far from autonomous

https://www.renascence.io/news/104630/ai-agents-overstate-their-results-and-remain-far-from-auton...
2•elorant•20m ago•0 comments

Medison – a small, free audio recorder and editor for Mac, Windows and Linux

https://www.medisonaudio.com
1•tylersrevenge•22m ago•0 comments

Preparing for the 2026 US Midterms

https://www.anthropic.com/news/2026-us-midterms
3•dgellow•22m ago•0 comments

Dogs Seem to Agree with Humans That 'Bouba' Is Round and 'Kiki' Is Spiky

https://www.sciencealert.com/dogs-seem-to-agree-with-humans-that-bouba-is-round-and-kiki-is-spiky
1•thunderbong•22m ago•0 comments

How to Convert Pipedream Workflows to FlowRunner, with Real Examples

https://flowrunner.ai/blog/pipedream-to-flowrunner-conversions
1•markpiller•22m ago•0 comments

Show HN: AI Interview Index – which companies allow AI in coding interviews

https://aiinterviewindex.com
1•MoritzWall•22m ago•1 comments

NCMNext-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation

https://scalingintelligence.stanford.edu/blogs/ncm/
1•matt_d•23m ago•0 comments

44Net Connect: A Simpler Way to Use 44Net

1•ARDC_73•23m ago•0 comments