tech mumus (@techmumus) header
tech mumus (@techmumus) profile picture

Canal do YouTube sobre tecnologia e inteligência artificial

Brazil
Joined June 2021
Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)Photo by tech mumus (@techmumus)
I did my PhD on geological CO₂ storage, so I'm a tough audience for explainer videos. This week I asked Claude Opus 5.5 to make one. In a single conversation it produced a two-minute motion-graphics video covering the full chain: capture at industrial sources, compression to dense phase, transport by ship or pipeline, injection into a saline aquifer, plume migration under the caprock, the four trapping mechanisms and seismic monitoring. What impressed me most wasn't the visuals but the judgement. It showed residual trapping as snap-off in a pore network, dissolution as convective fingering, and the plume stacking beneath thin shale layers inside the reservoir, as at Sleipner. It also flagged which parts were illustrative rather than real data, and what it couldn't verify itself. Every frame and the soundtrack were generated in code, with the animation synced to the music. No stock footage, no editing software. It's the clearest CO₂ storage explainer I've seen, and a good preview of how these tools will change technical communication in our field. (I still checked the geology. Old habits.) #CCS #CO2Storage #AI #Claude
1
2
763
tech mumus retweeted
Humans have an average of 200-250 ms of latency when speaking to each other. This voice model is even faster: only 110 ms of latency! Open-weights ←You don't need to pay anyone to use it. 8B parameters ← Small and cheap to host and run. You can run it locally by cloning the Github repository. They published the instructions in the repository below. Open models keep getting stronger!
🤝 Paid partnership
Today, we’re excited to introduce Miso One, the most emotive voice model in the world. Miso One is an 8-billion-parameter text-to-speech model for highly expressive speech generation. It emotes like a human and responds faster than a human, with just 110 milliseconds of latency. We’ve open-sourced the model weights, with API access coming soon. Hear how Miso One sounds in the thread below.
16
55
3
896
212,925
tech mumus retweeted
--dangerously-skip-permissions
179
1,065
101
15,191
656,882
🚀 AgentTip: OpenAI assistants inside any Mac app. Type @idea brainstorm an onboarding flow, hit ⏎, and AgentTip swaps the trigger for the assistant’s reply—right where you were typing. No context-switch, no copy-paste. One-time payment $4.99. #MacApps #ChatGPT #AIProductivity
1
1
106
The era of AI app subscriptions is coming to an end… Thanks to the advent of lightweight edge AI models and frameworks like cactus - developed by @RomanShemet (backed by @ycombinator) AI apps can run 100% on the phone My offline Plant ID app is proof: 👉 apps.apple.com/br/app/plant-… No more cloud fees. No more monthly charges. Just fast, private, one-time apps.
2
2
6
373
I've spent 200+ hours Vibe Coding games and apps. It's insane what you can build with just your voice and AI...ANYONE can do it. Here's everything I learned about Vibe Coding: 🧵
33
84
10
932
171,792
After many hours of scrutinizing humor in LLM outputs, this one by Claude 3.7 is the funniest by far.
LOL!! Claude (via Cursor) randomly tried to update the model of my feature from OpenAI to Claude 🤯 (my request was totally unrelated)
112
254
19
7,130
508,028
🔥🔥 CrewAI 0.22.5 is live! 💖Appreciate retweets 💖 📈 1.4M+ crews were run last week ⚒️ new CLI `crewai create project_name` 🤖 YAML for agents & tasks 🐛 Bug fixes ✨Introducing: crewAI+ our business-tier platform + our design partnership program crewai.com
42
104
20
422
58,607