Meet SuperPDF ⚡️
The cleanest way to scan, read, merge, compress & manage PDFs on iPhone.
No clutter. No confusing UI. Just super simple PDF tools that work.
Download SuperPDF now - https://t.co/elA9MGCZvg
SuperPDF — scan, edit, sign & fill PDFs. no subscription, no internet needed.
live on the App Store now.
best things get built with a friend at 2am.
https://t.co/elA9MGCZvg
Turn any PDF into something actually useful.
✨ Merge
✨ Compress
✨ Convert
✨ Fast. Clean. No nonsense.
Meet SuperPDF — the simplest PDF app you’ll actually enjoy using.
#SuperPDF#Productivity#iOSApps
I've been waiting for articles that point out the existential threat that AI coding technologies pose to the Indian tech sector. Here's one. I fear that the social and cultural—and hence political—consequences in India could be even worse than in the US.
https://t.co/2jBlCMlr1s
AI won’t entirely replace engineers, but it’s ratcheting up the pressure on the futures of the 1.5 million computer science grads India churns out each year https://t.co/cGXDcPO3nC
Everything’s on free tier for now, live in few hours.
•.tech domain: got free older domain
•DB + Auth: @supabase free tier
•Deployment: @vercel hobby
This weekend I tried @opencode and honestly it feels smooth. Will be trying more.
Try: https://t.co/i24BwrryXm
https://t.co/PtXfNyX0r7
Ask once. Multiple AI models answer, critique each other anonymously,then a synthesizer creates the optimal response — instantly.
#llm#openai#Anthropic#ai
We’ve released what we believe is the first of its kind open-source dataset pairing human facial expressions with actuator control data from a physically actuated humanoid face.
#AI#Robotics#OpenSource#HumanoidRobots#Infosys#HuggingFace
https://t.co/ECvWY6cMw5
Toward Greater Transparency: By visualizing what’s happening behind the scenes, we’re better able to catch errors like hallucinations, debug odd mistakes, and create LVLMs that are more trustworthy when handling critical tasks.
https://t.co/Yvz0dAJf9x
Accuracy vs. Grounding: These heatmaps reveal that even correct answers can come from focusing on the “wrong” region. Conversely, a model might lock onto the correct spot but still misinterpret what it sees.