“The greatest weapon against stress is our ability to choose one thought over another. Instead of worrying about what you cannot control, focus your energy on what you can create.”

7 LLM generation parameters, explained visually:

Kimi K2-0905 update 🚀 - Enhanced coding capabilities, esp. front-end & tool-calling - Context length extended to 256k tokens - Improved integration with various agent scaffolds (e.g., Claude Code, Roo Code, etc) 🔗 Weights & code: huggingface.co/moonshotai/Kim… 💬 Chat with new Kimi

CUDA 13.0 just dropped. I compressed their 26 page pdf into a thread:

5 Photography Tricks sc ig/andyyongfilms

This file was identified like #QuasarRAT catched my attention in the way how it tries to handle anti-reversing through the execution of hundreds "taskkill.exe" 🤔 taskkill /f /im procmon64a.exe taskkill /f /im Wireshark.exe taskkill /f /im x64dbg.exe app.any.run/tasks/e608fed3…

🚨 A critical SAP S/4HANA flaw (CVE-2025-42957, CVSS 9.9) is under active attack. Hackers only need a low-privilege account to seize full control—create superusers, steal data, even drop ransomware. Patch now before it’s too late. Details here → thehackernews.com/2025/09/sap-s4…

The evolution of modern RAG Architectures. 👇 In general there are so many variations you can choose from but it all eventually converges to some flavour of Agentic RAG. 𝗡𝗮𝗶𝘃𝗲 𝗥𝗔𝗚: the regular “embed query -> retrieve -> inject retrieved context into prompt” type of

AI agents can finally talk to your frontend! The AG-UI Protocol bridges the critical gap between AI agents and frontend apps, making human-agent collaboration seamless. MCP: Agents to tools A2A: Agents to agents AG-UI: Agents to users 100% open-source.

Big news: Introducing Qwen3-Max-Preview (Instruct) — our biggest model yet, with over 1 trillion parameters! 🚀 Now available via Qwen Chat & Alibaba Cloud API. Benchmarks show it beats our previous best, Qwen3-235B-A22B-2507. Internal tests + early user feedback confirm:

Fundamentals of LLMs Architecture (Transformer-Based Models) → Input Embeddings Words are first converted into numerical vectors. → Positional Encoding Since Transformers don’t have recurrence, positional encodings are added to embeddings to retain sequence order. →

Let's generate our own LLM fine-tuning dataset (100% local):

the benchmark to end all benchmarks

Not all Azure logs are created equal. Some stick around for 90 days, some for years, and others vanish unless you export them. I broke it down in Demystifying Log Retention in Azure. shankuehn.io/post/demystify… #monitorallthethings #log #retention #azure #cheat #sheet

Salesloft Drift Breach Tracker: Companies affected by the widespread Salesloft Drift OAuth token compromise that targeted Salesforce customer instances Link: driftbreach.com

Google presents an AI system to write expert-level scientific software. Using LLMs + tree search, it invented novel methods in bioinformatics, epidemiology, geospatial analysis & more, often surpassing human SOTA. (1/4)

A group calling itself Dark Samural, allegedly linked to OceanLotus, conducted attacks in Pakistan using MSC files combined with GrimResource techniques to deliver malicious payloads, but analysis revealed it was a false-flag operation by Patchwork ctfiot.com/269659.html

Building even a simple 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗴𝗿𝗮𝗱𝗲 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 𝗔𝘂𝗴𝗺𝗲𝗻𝘁𝗲𝗱 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 (𝗥𝗔𝗚) 𝗯𝗮𝘀𝗲𝗱 𝗔𝗜 𝘀𝘆𝘀𝘁𝗲𝗺 is a challenging task. Read until the end to understand why 👇 Here are some of the moving parts in the RAG based systems that

Highly recommend reading this great article by Jamf😍 jamf.com/blog/chillyhel… Some IOCs (see the write-up for the full list 😉): - Mach-O: 6a144aa70128ddb6be28b39f0c1c3c57d3bf2438 - Team IDs: R868N47FV5, F645668Q3H - IPs: 93[.]88[.]75[.]252, 148[.]72[.]172[.]53

ring ring... who is it? a bucket of sand to bury your head in. Join us today on our journey through FreePBX's CVE-2025-57819, as we chain unusual behaviour to demonstrate the auth bypass, to SQLi, to RCE, used in-the-wild. labs.watchtowr.com/you-already-ha…