Hudan Studiawan (@studiawan) header
Hudan Studiawan (@studiawan) profile picture

.

Surabaya, Indonesia
Joined June 2009
Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)Photo by Hudan Studiawan (@studiawan)
My new book is out! Drone and UAV Forensics: A Hands-On Approach link.springer.com/book/10.10…
907
Hudan Studiawan retweeted
api indonesia udah masuk hermes marketplace yg pake hermes agent tinggal hermes plugins install api-indonesia, langsung dapet data wilayah sampe desa, kode pos, libur nasional, gempa bmkg, kurs BI, kbbi dll. Rp2 per call hermes-agent.nousresearch.co…
22
102
3
939
24,001
Hudan Studiawan retweeted
Si quieres ejecutar modelos de IA en local, esta herramienta te va a ahorrar bastante tiempo. Se llama llmfit y analiza el hardware de tu equipo para decirte qué modelos tienen más sentido según tu CPU, GPU y memoria disponible. Así funciona: ✓ Analiza automáticamente tu hardware ✓ Compara tus recursos con distintos modelos ✓ Te muestra cuáles puedes ejecutar mejor ✓ Genera un ranking con las opciones más recomendadas para tu máquina La idea es dejar de probar modelos a ciegas y saber desde el principio cuáles encajan mejor con tu equipo. Es completamente gratuita y open-source. Link 👇
24
135
3
922
45,914
Hudan Studiawan retweeted
Run a 2.78 TRILLION parameter AI model with just 8GB of RAM. No GPU needed 😳 Meet kimi-k3-in-c A FREE open-source engine that runs the full Kimi K3 model on a single CPU. What makes it crazy? > Full 2.78T parameter model > Runs with just 8.24GB RAM > No GPU, PyTorch or CUDA required > Only 176KB of C code > Same output as a 128GB machine, but slower > Works on Linux, macOS and Windows How to run it: 1. Clone the repo github.com/FareedKhan-dev/ki… 2. Run make 3. Run ./scripts/k3-doctor.sh to check if your PC can handle it. 4. Download the model checkpoint and choose a RAM preset - It's slower and needs plenty of SSD storage, but running a trillion-parameter AI on 8GB RAM is WILD. Would you try this on your PC
Run multiple AI coding agents at once without conflicts 😳 Worktrunk - Manage Git worktrees as easily as branches - Create worktrees and launch Claude Code or Codex with one command - Shared caches and easier merges 8.3K GitHub stars
13
27
1
254
21,181
Hudan Studiawan retweeted
وداعاً لفلاشات الـ USB وصعوبة حرق توزيعات لينكس! 🚫💻 مطور يبتكر أداة hop تطبيق ويندوز خارق يتيح لك عمل Dual Boot لنظام لينكس بدون فلاشة USB، بدون تعديل يدوي، وبدون حوسبة معقدة! 🤯🚀 كيف تعمل الأداة ببساطة؟ 1️⃣ اختر التوزيعة التي تريدها والمساحة التي تحتاجها. 2️⃣التطبيق يقوم بتحميل الـ ISO تلقائياً. 3️⃣ يصغر قرص C: تلقائياً عند الحاجة. 4️⃣ ينسخ مثبت التوزيعة إلى قسم FAT32 صغير ويعيد تشغيل الجهاز لتبدأ التثبيت مباشرة! لا داعي لـ Rufus ولا إعداد قوائم الإقلاع اليدوية المعقدة بعد اليوم. 🔗 رابط المشروع على GitHub:👇👇👇👌 #Linux #Windows #DualBoot #OpenSource
4
24
214
13,882
Hudan Studiawan retweeted
写论文用在线 LaTeX 编辑器,合作者一多、想翻完整的历史版本,就得升级付费套餐。 最近看到 Oleafly 这个开源编辑器,LaTeX、Typst、Markdown 都能写,多人实时协作、版本历史、Zotero 引用这些都免费,无需账号。 和合作者改同一篇时,能看到每个人的光标和评论,实验室自己的服务器一条命令就能搭好。离线改的内容联网后自动合并。 编译在自己电脑上跑,装好就能出 PDF,点 PDF 能跳回源码对应的位置。编译出错时,上一版能用的 PDF 还留在屏幕上。 GitHub:github.com/Oleafly/Oleafly 自带 1700 多个模板,论文、毕业论文、幻灯片、简历都有。Claude Code、Codex 等 14 个 Agent 能直接在项目里干活,每处改动都能单独确认或撤销。 投稿前的检查我觉得最实用,引用断了、图片缺了、字体没嵌入,还有盲审时会暴露作者身份的地方,都会提前标出来。
15
61
7
447
28,142
耶鲁计算机博士生开源了自己的论文画图脚本 figures4papers 柱状、雷达、折线、3D球体、概念图都支持,8个论文项目的完整画图方案都在里面,改改数据就能直接用 GitHub:github.com/ChenLiu-1996/figu… --- 规范细到不行——去掉上下坐标轴线、柱子加黑边、配色按语义走(蓝色自己的方法、绿色改进、红色对比基线),300 DPI 导出 直接改数据最方便,不想手动改也行,把仓库接进 Cursor 或 Claude,按 SKILL.md 的规范让 AI 帮写脚本,README 里有现成的提示词 --- 论文快deadline了还在纠结怎么出图的可以收藏备用,用顶会作者的审美标准往往不会翻车
5
114
3
812
37,702
Hudan Studiawan retweeted
绝了!NVIDIA最新开源了2,575 个高规格 3D 资产,整整472GB!! 从日常用品到工业设备,里面收录了不少现实环境里的物体。 一揽子,全部放进了 Hugging Face。 NVIDIA SimReady Assets。 绝不单纯是一个“免费建模素材库”!! NVIDIA给每个3D 资产都带上了物理属性和语义信息,比如有多重,什么材质,重心在哪,用在什么场景等等,全都写进了参数里。 明摆着,就是用来让AI学习的! 学习怎么和真实的物理世界互动! 传统的3D资产在AI眼中,就是个有贴图的“视觉资产”,NVIDIA想做的是,把“视觉资产”变成AI可以理解的“物理资产”! 所以,这批资产的价值可想而知! 至少能够用到这些场景上! 机器人抓取和搬运训练,先在虚拟环境里练习拿取、摆放物体; 搭建工厂、仓库的数字孪生,测试设备和物体的布局; 配合合成数据工具,改变光照、视角、遮挡和物体位置,生成带标注的训练数据。 大家都在卷文字、代码与图像生成时,NVIDIA 开始为Physical AI(物理 AI) 铺设地基了。 不过,对大部分人来说,也是一个相当全面的3D资产素材库! 链接在下面,必须收藏了!
51
407
35
3,732
212,290
Hudan Studiawan retweeted
Added a new skill: /teardown It creates a technical illustration of any object and lets you customize individual parts, colors and annotations. You can then export the result as an SVG or PNG. npx skills add jakubkrehel/skills
99
282
46
6,090
247,346
Hudan Studiawan retweeted
Hardware looks scary until you break it down into small steps. So I built an app that takes you through Arduino one circuit at a time. Made with Opus 5.5 learn-arduino-red.vercel.app
333
1,554
160
14,527
901,497
Hudan Studiawan retweeted
Anthropic released code-modernization v1.0.0, a Claude Code plugin for modernizing legacy codebases Save this post if you have a legacy project worth modernizing What I find interesting is that it doesn’t just rewrite code. It tries to prove the new implementation behaves like the original Here’s how: - Analyzes the legacy system and extracts business rules - Generates a modernization plan for your approval - Rewrites or upgrades the code, even across languages - Runs the old and new implementations with identical inputs and compares outputs - Generates additional edge cases to test behavior - Introduces deliberate bugs to verify tests actually catch them - Produces a verification report with PROVEN, PARTLY PROVEN, or NOT PROVEN results It also keeps the original code untouched and requires human approval at critical steps I think this is the right direction for AI-assisted modernization, generating new code is one thing Proving that you haven’t broken existing behavior is a completely different challenge Install:
/plugin install code-modernization@claude-plugins-official Repo:
github.com/anthropics/claude…
71
184
32
2,341
188,759
Hudan Studiawan retweeted
我愿称之为史上最强 skills skills.wheresryan.sh/anatomy
150
738
38
6,282
433,498
Hudan Studiawan retweeted
You can literally type a sentence and watch a Transformer process it. This interactive explainer runs a GPT-style model in the browser and lets you follow the whole path: tokenization → embeddings → Q/K/V → attention → MLP → logits → next-token probabilities You can inspect attention maps, intermediate values, and change temperature, top-k and top-p to see how generation changes. If Transformers still feel like boxes connected by arrows, this is worth opening. poloclub.github.io/transform…
37
370
25
2,387
136,133
Hudan Studiawan retweeted
You'll enjoy the asm visualizer asm.diveintosystems.org/
Book: x86-64 Assembly Language Programming with Ubuntu egr.unlv.edu/~ed/assembly64.…
1
60
420
12,338
Hudan Studiawan retweeted
逆向工程的玩法彻底变了… 🤯 刚刚试了这个 GitHub 开源神仙项目:x64dbg-mcp-server。 简单来说:它把 Windows 最老牌的逆向神器 x64dbg 做成了 MCP 服务。 这意味着什么?你现在可以用 Claude / AI 直接接管调试器。 以前逆向分析: 手动下断点 ➡️ F7/F8 单步步过 ➡️ 肉眼找加密函数 ➡️ 翻内存栈 现在: 直接对 Claude 说一句: “载入恶意样本,在入口点断下,步进5步,提取内存解密后的字符串。” AI 自己调了 8 个 API,秒级把寄存器和反汇编分析全部吐出来… ✨ 亮点: 1. 80 多个调试控制 API(内存读写、脱壳OEP检测、断点、寄存器) 2. 纯 Zig 写的,零环境依赖,dll 丢进去就能跑 3. 不需要 Python 或庞大框架,直接支持 SSE/HTTP 以后打 CTF 逆向或者做外壳分析,真的要变成“聊天解题”了吗?
22
156
3
1,003
58,127
Hudan Studiawan retweeted
I wanted to visualize what actually happens inside an LLM. So I made this. It got a little out of hand. Built with Claude Opus 5.5 + GPT-6.1 Sol. (Sound on)
21
57
10
373
24,975
Hudan Studiawan retweeted
I curated 125 free math textbooks, written by the mathematicians who teach the courses and shared on their own university pages. No sign-up, no paywall, no pirated copies. From high school algebra to graduate analysis. 👇 abakcus.com/book-lists/free-…
20
412
16
1,868
104,303