Most of our explaining at Hobglobin now happens in HTML pages, not slides. Claude writes them in minutes, then they get passed around the team.
The weak spot: an HTML file can run code, and we were opening each other's without a second thought. So we built a small utility.
2
53
It turns any HTML page into one file that looks the same but can't do anything: no scripts, no trackers, no forms. Anyone can check a file before opening it.
It runs as a Claude skill, so explainers come out clean and checked before anyone shares them.
1
1
31
Open source as inerthtml:
github.com/Hobglobin/inertht…
Check an HTML file in your browser (nothing is uploaded):
hobglobin.github.io/inerthtm…
21
Sambit Mahapatra | Construction Compliance AI retweeted
We're giving away a CONTROL Resonant custom wrapped NVIDIA GeForce RTX 5080 to celebrate its official release with #RTXON.
To enter:
🟢 Share this post
🟢 Comment #RTXON
T&Cs: nvda.ws/4z24Gcq
Sonnet 5.5 is better than Opus 5.5 in terminal bench 😳
Anthropic is adamant on getting all good wills back
Replying to @claudeai
Sonnet 5.5 improves on Sonnet 5 across benchmarks, in some cases dramatically.
It’s a faster, lower-cost complement to Claude Opus 5.5, strongest at well-scoped everyday tasks, fixing bugs, and creating polished documents, slides, and spreadsheets.
22
NVIDIA released a fast speaker-labeling model today: Nemotron 3 Diarization. Like all of the Nemotron family from @NVIDIA_AI, this is an open weights model, so you can grab the weights from @huggingface and use the model in the cloud or on your local device.
Nemotron 3 Diarization is a useful building block for voice agents and realtime AI applications that need to handle speech from multiple people at the same time. It handles overlapping speech well, can identify up to eight speakers, and pairs with any transcription model you're using in your multi-modal pipelines.
@jonptaylor built a demo (code below) and recorded a technical deep-dive showing how to use this diarization model together with a speech-to-text model.
Jon's code also uses Jev for super-accurate confirmation of critical voice input in a noisy/multi-speaker environment.
A 7B model above nano banana, let that sink in! huggingface.co/Qwen/Qwen-Ima…
118
Fascinating!
Replying to @CompleteSkeptic
Extraordinary claims require extraordinary evidence so check out our release blog for more technical info: typesafe.ai/blog/introducing…
Join our waitlist for early access: typesafe.ai/
Have technical chats and meme with us on Discord (rumors are good memers skip the line): discord.gg/WUujKYBp8s
We are so pumped this is in the hands of developers now, and we are just getting started!!!
41
Starting in 5, waiting to be surprised and feel more existential crisis teapawt.net/i/broadcasts/1AxRnZbVp…
13
Anthropic's threat report is impressive. they caught real misuse. Respect.
But to catch it, they read and retained user conversations. Including yours, potentially: one false safety flag and real humans are reviewing your prompts.
Enterprise buyers: "we have an enterprise plan" is not an architecture. If you're regulated, run your own models. Open Wight models maybe 6 months behind the frontier now.
Nothing to retain = nothing to leak.
10
Sambit Mahapatra | Construction Compliance AI retweeted
AGENT HARNESSES ARE THE NEW GPT WRAPPERS
What exactly is an agent harness? A harness does 4 things:
1. Runs the model in a loop so it keeps working step after step instead of answering once and stopping.
2. Gives it hands to read files, call tools, open portals, and run code.
3. Manages its memory so hour three of a job still knows what happened on hour one.
4. Enforces the rules about what it can touch and when it has to stop and ask a human.
WHY IT'S INTERESTING
- The market is MEGA. I think about it like a wrapper let you sell software ($800B market) but a harness lets you sell the work ($5T+ market)!
- It doesn't depend on any one company's model. The knowledge about the job lives in the harness, so you can run GPT today, Claude next month, an open model like Google Gemma, Qwen, Deepseek etc on your own machine after that, and it keeps working. A wrapper was one model doing everything and a harness is a router.
- It gets better the more you use it. See, every correction a human makes becomes a rule the harness keeps, so what you have after 500 jobs is a completely different product than what you had after 5. Wrappers only got better when OpenAI got better.
- You can charge for finished work! The harness knows when a job is done, so you can price per claim, per filing, per review, per resolved exception, or per closed month. This helps compete with SaaS!
- Most jobs are just the same handful of decisions, repeated, using the same few tools. That's true for most of the 800+ occupations out there, which is why almost all of them could have a harness built for them. Lots of space for startups to be building.
-The frontier labs won't come for these. OpenAI is not going to learn how a freight claim gets denied in Rotterdam or which prior auth your specific payer rejects. Those markets are too small for them and the knowledge only exists inside the job. They'll keep making the models better, which just makes your harness better which is cool.
- OH, AND really important, you can ACTUALLY build one now. OpenAI's Agents API, Claude's managed agents, Vercel's eve, LangChain harness framework etc. They all make it way easier to build agent harnesses.
Agent harnesses truly are the new GPT wrappers.
This is just the beginning.
Time to download and store the llm open weights and go finetuned if needed. Else AI innovation will be limited by the access
20
Let’s just ban open source ai and the frontier AI will pivot from being dangerous to being safe again. Then no need of slowing down the AI
#AIsafety
19
Sambit Mahapatra | Construction Compliance AI retweeted
Best local AI models to run on your DGX Sparks ✨
1x DGX Spark - Qwen3.8-Flash-Next
github.com/MiaAI-Lab/Qwen3.8…
------------------
2x DGX Sparks
DeepSeek v4 Flash
github.com/MiaAI-Lab/DeepSee…
GLM 5.3 Flash
github.com/MiaAI-Lab/GLM-5.3…
Qwen3.8 Flash
github.com/MiaAI-Lab/Qwen3.8…
------------------
3x DGX Sparks - DeepSeek v4.1 Flash TP3
github.com/MiaAI-Lab/DeepSee…
------------------
4x DGX Sparks - DeepSeek v4.1 Flash TP4
github.com/MiaAI-Lab/DeepSee…
















