frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: Fixing LLM memory degradation in long coding sessions

https://github.com/robertomisuraca-blip/LLM-Entropy-Fix-Protocol
2•robertomisuraca•1h ago
Long-session LLM memory degradation (entropy) is the silent killer of complex coding projects. Models like Gemini, GPT-4, and Claude all suffer from it, leading to hallucinations and lost context.

I've developed an open-source protocol that temporarily "fixes" this issue by structuring the dialogue. It's not the final architectural solution, but it’s a proven patch for developers working right now.

Looking for feedback from the community on how we can solve this structurally. https://github.com/robertomisuraca-blip/LLM-Entropy-Fix-Prot...

Show HN: I got tired of doing SEO work so I automated it

https://blogseo.io
1•vincejos•24s ago•0 comments

How to Deal with AI Restrictions When Upscaling Images

https://ivanca.github.io/programming/2025/11/26/bypass-ai-upscalint-restrictions/
1•AmbroseBierce•1m ago•0 comments

Major LLMs fabricate evidence when presented with governance critiques

https://zenodo.org/records/17728992
3•mikeup91•10m ago•1 comments

New research highlights a shortage of male mentors for boys and young men

https://www.psypost.org/new-research-highlights-a-shortage-of-male-mentors-for-boys-and-young-men/
6•ashishgupta2209•17m ago•0 comments

Using Claude to create a bootable Forth OS

https://isene.org/2025/11/SimplicityOS.html
2•veltas•27m ago•0 comments

My Rebuttal to Simon Cowell's Rolling Stone Statement

https://twitter.com/katiewaissel24/status/1993828722500055299
2•mellosouls•28m ago•0 comments

Claude Opus 4.5, and why evaluating new LLMs is increasingly difficult

https://simonwillison.net/2025/Nov/24/claude-opus/
2•jonesn11•29m ago•0 comments

AuthenticImage – Detect AI-Generated / Fake / Edited Images (100 users on day 1)

https://authenticimage.site/
1•CrazyCompiler01•30m ago•1 comments

Finding a private self hosted Google Photos alternative that doesn't track me

https://alternativeto.net/news/2025/11/finding-a-private-self-hosted-google-photos-alternative-th...
1•elliot_a•31m ago•0 comments

The EU's Silent Assault on Human Dignity

https://twitter.com/yrschrade/status/1989718227690119560
2•nailer•32m ago•0 comments

Adventures in live booting Linux distributions (2014)

https://major.io/p/adventures-in-live-booting-linux-distributions/
1•transpute•36m ago•0 comments

Don't tell me RAG is easy

https://major.io/p/devconf-rag/
2•transpute•37m ago•0 comments

U.S. to Charge International Tourists $100 to Visit National Parks

https://skift.com/2025/11/26/u-s-to-charge-international-tourists-100-to-visit-national-parks/
1•decimalenough•37m ago•1 comments

Show HN: Pg_AI_query – AI powered SQL inside PostgreSQL

https://github.com/benodiwal/pg_ai_query
1•benodiwal•41m ago•0 comments

Leaked Celebrite Android Payload

https://donotsta.re/notice/B0etpuWyWytSdbRnTU
1•Aissen•42m ago•0 comments

Coffee Making

https://tldp.org/HOWTO/html_single/Coffee/
1•Lunar5227•43m ago•0 comments

Chinese businesses are transforming Mexico City's poshest neighborhood

https://restofworld.org/2025/chinese-tech-companies-mexico-city/
1•Brajeshwar•46m ago•0 comments

Founder-led sales, from the perspective of a Founder, leading sales

https://substack.com/inbox/post/180021176
1•mnbbrown•48m ago•0 comments

OpenDesk 1.10. Enhanced security architecture

https://www.opendesk.eu/en/blog/opendesk-1-10
2•doener•52m ago•0 comments

Apple Asks Indian Court to Block Antitrust Law Allowing $38B Fine

https://macmegasite.com/2025/11/26/apple-asks-indian-court-to-block-antitrust-law-allowing-38-bil...
2•Brajeshwar•53m ago•0 comments

Tell HN: Stall AI progress for the benefit of humanity

5•blutoot•55m ago•5 comments

Show HN: Whisper Money – A privacy‑first, E2E encrypted personal finance app

https://github.com/whisper-money/whisper-money
1•falcon_•59m ago•0 comments

Show HN: Flux 2–High-Fidelity Image Generation with Multi-Reference Consistency

https://www.fluxproai.net/?i=d1d5k
2•lu794377•1h ago•0 comments

Show HN: Fixing LLM memory degradation in long coding sessions

https://github.com/robertomisuraca-blip/LLM-Entropy-Fix-Protocol
2•robertomisuraca•1h ago•0 comments

Tell HN: DuckDuckGo doesn't have bangs for Chatbots like ChatGPT, Grok, Gemini

2•faebi•1h ago•1 comments

Record Club: indie social network for music lovers

https://record.club/
1•thwg•1h ago•1 comments

RL is more information inefficient than you thought

https://www.dwarkesh.com/p/bits-per-sample
2•cubefox•1h ago•0 comments

Show HN: I built a language learning movie app

https://kanjieight.vercel.app/
1•Mikecraft•1h ago•0 comments

How to Synthesize a House Loop

https://loopmaster.xyz/tutorials/how-to-synthesize-a-house-loop
3•stagas•1h ago•0 comments

Morgan Stanley Warns Oracle Credit Protection Nearing Record High – Above 2008

https://www.bloomberg.com/news/articles/2025-11-26/morgan-stanley-warns-oracle-credit-protection-...
5•zerosizedweasle•1h ago•1 comments