Raj Nakarja

Back when I was in my tiny lab, I used to frequently end up with headaches and feel fatigued. Lo and behold, it was simply a lack of fresh air! Thankfully that’s no longer an issue! docs.siliconwitchery.com/pages/blog/eas…

Had a nostalgia moment and decided to start blogging again! Back in the day I used to document PC builds and speakers, but now I figured it would be fun to document my IoT projects since I’m working a lot with LTM-M docs.siliconwitchery.com/pages/blog/

After months of work, I'm excited to share our first demo of the S2 & Superstack!

@chrypnotoad Interesting how our brain can still pick out an uncanniness to this if you’ve been an actual Minecraft player. It’s not far off though. I wonder if that uncanny feeling will be surpassed as the technology improves, or if we’ll become used to it and start to ignore it completely

lol Sora can just generate minecraft gameplay.

Deep dive about Frame hardware, firmware and design. As well as updates to our Noa AI companion twitter.com/i/spaces/1ZkKz…

We're teaming up with @brilliantlabsAR to integrate our real-time answers into Frame, their AI-enhanced glasses. You can ask about what you see, instantly — all in your line of sight.

What if your glasses gave you AI superpowers? This is Frame, a reinvention of eyewear for the AI era Learn more @ brilliant.xyz

@QVHenkel This is fantastic. Have you tried it with paste yet?

Another interpretation of flip-chip that I wanted to try since I got my new SLA printer😁I have 3D printed this small component holder as an assembly aid. The PCB with solder paste already applied is flipped onto the component holder, which has a fixed distance of 0.15mm to the PCB surface. The components stick to the paste and the whole stack can be turned over again and the holder lifted off. It works well without solder paste and the components simply fall onto the PCB perfectly aligned 😊

@larianstudios If you're going to do updates separately, at least let users choose if they want to stay on the same version or not. It's as if someone came to a gaming table, and took away one person's character sheet while the others were allowed to keep on playing

Mac users, we continue to work on Patch 5 for your platform and expect it to have it available soon along with this last hotfix. We'll keep you updated here.

Hotfix 12 is now live for Baldur's Gate 3. This update addresses crashes that were occurring in certain instances, such as starting a conversation with Cazador and joining with a second controller and activating split-screen in certain places. Read More: larian.club/HF12

Transparent AR will be the best interface for AI. Here's multimodal AI working on @brilliantlabsAR Monocle using LLaVA! And it's all open source (see 🧵👇) (1/)

ロボットの動作確認をしているときにディスプレイを見るのがわずらわしくなってきたのでMonocleを導入してみた。バッテリーとかモーターの状態を表示させていて結構快適。カメラがついているので「エラーが起きたときに自分が見ているロボットの様子を自動で撮影しておく」とかもできそう

@karpathy It could also be chaos 😂 but just as people discuss and negotiate to get things done in real life, these devices could do the same. Each finding some way to solve a problem without relying on plain old instructions to do so

@karpathy Devices could also start talking to each other directly. Cameras reaching out to IMUs in the system to get more motion data for image stabilization. All without host involvement, or specific programming/pairing of devices

With many 🧩 dropping recently, a more complete picture is emerging of LLMs not as a chatbot, but the kernel process of a new Operating System. E.g. today it orchestrates: - Input & Output across modalities (text, audio, vision) - Code interpreter, ability to write & run programs - Browser / internet access - Embeddings database for files and internal memory storage & retrieval A lot of computing concepts carry over. Currently we have single-threaded execution running at ~10Hz (tok/s) and enjoy looking at the assembly-level execution traces stream by. Concepts from computer security carry over, with attacks, defenses and emerging vulnerabilities. I also like the nearest neighbor analogy of "Operating System" because the industry is starting to shape up similar: Windows, OS X, and Linux <-> GPT, PaLM, Claude, and Llama/Mistral(?:)). An OS comes with default apps but has an app store. Most apps can be adapted to multiple platforms. TLDR looking at LLMs as chatbots is the same as looking at early computers as calculators. We're seeing an emergence of a whole new computing paradigm, and it is very early.

Nice summary of my recent chat with the folks on the @hackaday hackchat ☺️

Coming soon to a Monocle near you! 🧐 @brilliantlabsAR

@arturo182 Soldering iron controller on the left, and AR glasses motherboard on the right. Seems I always had a thing for narrow little boards 🤓

You! PCB Designer! Show me your first PCB design and your latest one!

Unboxing my craziest board layout yet. The 01005 pads are the tightest IPC spec + the exact tolerances of PCBWay’s pick and place machines 😅 youtu.be/2ezC-eDhUBo