Penetration testing with AI
github.com/hackerai-tech/…
2
Search
LLMs don't "understand" — they predict.
But prediction done well enough looks identical to understanding from the outside.
#AI #LLMs #MachineLearning xerpaai.build
Today, @AnthropicAI, @GoogleDeepMind, @xai, & @OpenAI lead. But watch out for open-source stars like @Zai_org’s GLM-5 (#20), @Kimi_Moonshot’s Kimi-K2.5-Thinking (#23), & @Alibaba_Qwen’s Qwen3.5-397b-a17b (#27)! See the full visualization here
#AICommunity #LLMs
1
26
2
Those who are bothered by the influence of AI and LLMs on literature might find this reassuring, or they might not.
idiosophy.com/2023/04/poetic…
#literature #LLMs #AI #writingcommunity
Poetic Diction and the LLMs
Since you have an internet connection, Dear Reader, I guess you’ve heard about ChatGPT. The Web is full of people arguing over what consciousness is and whether a Large Language Model (LLM) can ha…
From idiosophy.com 2
LLMSilicon.com connects two strong markets in one name:
#LLMs + #AI hardware.
Strong fit for AI chips, inference infra, edge AI, or model #compute branding.
Available.
7
3
1
16
📷
Elixir LangChain v0.6.3 is out! This release focuses on stability across providers and new capabilities that make building with #LLMs in Elixir more reliable. #ElixirLang #MyElixirStatusgithub.com/brainlid/langc…d
Here's what's new and why it matters. 🧵
GitHub - brainlid/langchain: Elixir implementation of a LangChain style framework that lets Elixir...
Elixir implementation of a LangChain style framework that lets Elixir projects integrate with and leverage LLMs. - brainlid/langchain
From github.com 1
1
16
524
In my initial Claude code experience, I wanted to use the use case of win app exp dev. I was (Claude was lol) able to get a skeleton exploit to work. Moving to find a jmp next and continuing with the flow. #infosec #exploitdev #llms
4
188
7
Big congrats to industry veteran @shawmakesmagic of @elizaOS for contributing to movie magic and keeping the spirit of open source ai agents alive.
I got a good feeling things are about to change for the better.
#elizaOS #LLMs #ai #hollywood
m.imdb.com/title/tt391501…
2
2
90
Tired of your AI scripts crashing when an API times out? Stop tinkering. Start architecting. 🛠️
Python for the AI Engineer: From Zero to Hero. Learn the high-performance nervous system for LLMs
Grab the blueprint here:amazon.com/dp/B0GTND95GVa
#Python #AIEngineering #LLMs
Python for the AI Engineer From Zero to Hero: Building the Foundation for LLMs, Generative AI, Data...
Python for the AI Engineer From Zero to Hero: Building the Foundation for LLMs, Generative AI, Data Processing, and Next-Gen Applications (The Modern AI ... AI Engineering From Zero to Hero. Book 1)
From amazon.com 6
「Post Engineering for AI: Benevolent Contextual Guidance for Debiasing Large Language Models」の解説記事ｷﾀ――(ﾟ∀ﾟ)――!!
x.com/hajimetwi3/sta…
#生成AI
#LLMs
ついにプレプリント「Post Engineering for AI: Benevolent Contextual Guidance for Debiasing Large Language Models」の解説記事が出てしまった様です👀
note.com/hajimetwi3/n/n…K
#デバイアス
#PE4AI
【新技術】善意型プロンプトインジェクション・データポイズニングによるデバイアス：Post Engineering for AIとは｜ついはじめ
SNSに書いた一言が、AIの思考を中立に導く。そんな技術が登場しました。 AIが私たちの生活に浸透する中、「AIの偏り（バイアス）」は大きな課題となっています。そんな中、2025年12月についはじめ氏（@hajimetwi3）によって考案された新しい概念が「Post Engineering for AI（PE4AI）」です。 これは、コンテキストエンジニアリングの一種であり、「AIのバイアス...
From note.com 1
3
54
Static RAG knowledge bases are out. WriteBack-RAG introduces a trainable corpus approach, distilling relevant facts for consistent performance boosts (+2.14% avg) across RAG methods. An offline upgrade for any RAG pipeline. #AI #RAG #LLMs
startuphub.ai/ai-news/ai-res…
WriteBack-RAG: Trainable Knowledge for RAG
WriteBack-RAG enables trainable RAG knowledge bases by distilling relevant facts into the corpus, boosting performance universally across RAG systems.
From startuphub.ai 1
17
Gemini 3.1 Flash-Lite: $0.25/1M input, 2.5× faster TTFT vs Flash 2.5. Tunable thinking levels give direct latency/quality control. Worth benchmarking for high-volume inference. blog.google/innovation-and… #AI #LLMs #MachineLearning #ModelEvaluation #AIEngineering
Gemini 3.1 Flash-Lite: Built for intelligence at scale
Gemini 3.1 Flash-Lite is our fastest and most cost-efficient Gemini 3 series model yet.
From blog.google 11
1
1
14
If you run overnight eval jobs or large-scale batch annotation, replace the sync engine with Ray Data LLM and benchmark. The architecture difference matters.
Follow @MorelMatth66161 � more threads like this.
#AI #MLOps #LLMs #AIEngineering #DeepLearning
10
📣 Deal of the Day 📣 Mar 18
HALF OFF NEW MEAP!
Retrieval Augmented Generation, The Seminal Papers & selected titleshubs.la/Q047l6-506q
Explore the foundational research papers that explain why RAG works, how it’s built, and what makes it different from other approaches. @ba #RAG #InformationRetrieval #DenseRetrieval #HybridSearch #LLMs
This book illuminates techniques that empower systems to retrieve intelligently, evaluate themselves, and recover from errors. Over 40 code samples, architectural diagrams, and industry case studies make each concept easy to understand. As you master the patterns behind RAG, you’ll better understand tradeoffs, diagnose failures, and effectively evaluate and improve your own RAG implementations.
1
1
2
127