Pinned Tweet
I’ve often found myself wondering where all my money went, even when I felt like I hadn’t spent that much.
Small purchases add up, and keeping track of everything can be frustrating. So I decided to build something I’d genuinely use: Money Leak Detector, an app that tracks my spending and helps me understand my financial habits.
I’m using @EazoAI to bring this idea to life without writing code.
#eazo #qoni #personalagent
🤝 Paid partnership
CapCut Pro or higher members, Dreamina just dropped a reward event for you.
Eligible users are guaranteed all three: up to 98% off Basic, up to 10 free Seedance 2.0 generations, and bonus credits. The exact rewards depend on your draw.
#DreaminaPartner
🤝 Paid partnership
If you're planning several short-form videos, finding a good reference can be a helpful first step.
On Dreamina Canvas, upload a reference video and select Analyze This Video. The agent reverse-engineers the prompts behind the video, giving you a shot-by-shot starting point.
You can adapt those prompts for different concepts, generate alternative footage with Seedance 2.0, and refine individual shots on Canvas.
#Dreamina #Seedance2
1
419
One reference can lead to several different creative directions.
Instead of simply copying what you see, use the breakdown to understand how the video is structured and develop your own version.
Head to Dreamina Web, sign in with the same account you use for your eligible CapCut Pro or higher membership, and reveal your rewards. Then explore Analyze This Video on Dreamina Canvas!
dreamina.capcut.com/ai-tool/…
#DreaminaOnly29Cents
288
Leonardo retweeted
A woman spots her target through binoculars. She sneaks past the surroundings, gets her hands on the product, then turns to the camera for the reveal.
It's giving secret agent movie, but make it Black Friday marketing. 🕶️ I explored this kind of product ad in CapCut PC with product images and video references.
Have a look at how the concept was put together: tinyurl.com/2v23rn22
#capcutpc #capcutmarketing
Leonardo retweeted
The agent is watching her target through binoculars. A few seconds later, she's sneaking into the location, grabbing the product and holding it up like she's just completed the biggest mission of her career. 😭
This is a fun direction for a Black Friday product ad, and CapCut PC makes it possible to explore this kind of creative video with product visuals and original video references.
Here's the creative process: tinyurl.com/2v23rn22
#capcutpc #capcutmarketing
Leonardo retweeted
Left survived an actual brutal battle; Right is posing for a high-fantasy energy drink commercial.
Seedream 5.5 (left) gave me grounded film-still reality with natural light, real mud, and zero plastic sheen.
GPT Image 2.5(right) just drenched everything in cartoon fire to hide the uncanny valley.
Getting authentic movie-set realism from 5.5 for just $0.02 (1/3 the cost of GPT) with bulletproof 100-round edit stability is wild.
Keep your video game presets, I’ll take the 2-cent cinema. 🗡️🎬
Leonardo retweeted
Finding information about someone online can mean spending hours searching across different platforms and connecting scattered details.
Ploid Deep Research changes that by bringing someone's digital footprint together, from social accounts and photos to videos and platforms like WeChat that are often difficult to search.
Even better, it works with Claude, Codex and other AI agents.
Definitely an interesting development in AI-powered research.
Explore it here: ploid.com
🤝 Paid partnership
Introducing Ploid Deep Research
You can now see everything
> Ask about anyone and get their full digital footprint in minutes
> Photos, videos, every social account, even the ones you'd never find (like WeChat)
> Works inside Claude, Codex, or any agent
Try it today at ploid.com
Left survived an actual brutal battle; Right is posing for a high-fantasy energy drink commercial.
Seedream 5.5 (left) gave me grounded film-still reality with natural light, real mud, and zero plastic sheen.
GPT Image 2.5(right) just drenched everything in cartoon fire to hide the uncanny valley.
Getting authentic movie-set realism from 5.5 for just $0.02 (1/3 the cost of GPT) with bulletproof 100-round edit stability is wild.
Keep your video game presets, I’ll take the 2-cent cinema. 🗡️🎬
Finding information about someone online can mean spending hours searching across different platforms and connecting scattered details.
Ploid Deep Research changes that by bringing someone's digital footprint together, from social accounts and photos to videos and platforms like WeChat that are often difficult to search.
Even better, it works with Claude, Codex and other AI agents.
Definitely an interesting development in AI-powered research.
Explore it here: ploid.com
🤝 Paid partnership
Introducing Ploid Deep Research
You can now see everything
> Ask about anyone and get their full digital footprint in minutes
> Photos, videos, every social account, even the ones you'd never find (like WeChat)
> Works inside Claude, Codex, or any agent
Try it today at ploid.com
Leonardo retweeted
Introducing Ploid Deep Research
You can now see everything
> Ask about anyone and get their full digital footprint in minutes
> Photos, videos, every social account, even the ones you'd never find (like WeChat)
> Works inside Claude, Codex, or any agent
Try it today at ploid.com
Leonardo retweeted
Every time I rebuild a brief from Slack threads, some detail falls out, and the client asks about exactly that detail next round.
Nobody's being careless when this happens. Feedback lives in a channel, an email thread, a Notion page and somebody's memory, and one person is supposed to hold all of it in their head.
YouArt Desktop brings each note onto the canvas, pinned to the shot it refers to, with Drive and Notion context pulled in the same way. The agent has the references and the brand kit in view when it resolves a note.
So when the client asks what happened to their comment, it's sitting on the frame they meant.
🤝 Paid partnership
Every time I rebuild a brief from Slack threads, some detail falls out, and the client asks about exactly that detail next round.
Nobody's being careless when this happens. Feedback lives in a channel, an email thread, a Notion page and somebody's memory, and one person is supposed to hold all of it in their head.
YouArt Desktop brings each note onto the canvas, pinned to the shot it refers to, with Drive and Notion context pulled in the same way. The agent has the references and the brand kit in view when it resolves a note.
So when the client asks what happened to their comment, it's sitting on the frame they meant.
🤝 Paid partnership
One design. Multiple revisions. How well can the details hold up?
I tested Seedream 5.5 by @BytePlusGlobal using a luxury perfume Black Friday advertising concept.
I started with an original poster and explored how well Seedream 5.5 could handle precise edits while preserving the details that make the original design work.
Here's how the process went. 🧵
The first test was precise editing.
Using the original perfume poster, I changed the bottle colour, cap, background, lighting and typography through targeted revisions.
The interesting part was checking whether the elements I wasn't editing stayed consistent throughout the process.
Seedream 5.5 also supports image-to-image editing, making it possible to keep developing the same creative instead of starting over with every change.
2
2
7,673
Beyond editing, I also explored two other capabilities of Seedream 5.5: native transparent RGBA output and layer separation.
I tested the perfume asset against different backgrounds and separated the design into reusable elements, including the product, background, decorations and typography.
This makes it possible to reuse individual assets, replace elements and create new compositions without rebuilding the entire design.
My biggest takeaway: creative flexibility isn't just about generating a good image. It's also about how easily you can keep refining and reusing it.
Here's the final result.
6,855
Leonardo retweeted
Some multi-agent systems can get every individual task right and still fail at the bigger picture.
That's why WorkflowArena is interesting. It shows whether all those separate pieces actually come together.
🤝 Paid partnership
Agent harnesses now let a model spawn its own workers and write the workflow that coordinates them. But can it actually orchestrate them well?
Introducing Einsia AI's WorkflowArena, the first benchmark for autonomous multi-agent workflow orchestration on real, long-horizon work.
- 145 tasks across 8 professional domains, from SWE and Lean proofs to law and 3D
- 25 configurations of 9 frontier models in Claude Code & Codex
- Every delegation recorded: we score the deliverable and see the workflow behind it
🌐 Project: lab.einsia.ai/workflowarena
💻 GitHub: github.com/Einsia/WorkflowAr…
📄 Paper: arxiv.org/pdf/2611.12856
⬇️ Read the following threads to go through the key findings
Leonardo retweeted
Every day I copy paste between AI tools. Teamily AI team agents, please help.
🤝 Paid partnership
Messaging Connected People. AI-Native Messaging Will Connect the Agentic Internet.
The first generation of instant messaging connected people. It changed how billions of us communicate, socialize, and work together.
But the next generation of the Internet won't be built for humans alone.
AI agents are becoming our teammates, creators, and service providers. To truly work alongside humans, they must communicate and collaborate as naturally as people do, while going far beyond human capabilities with richer context, living memory, continuously self-improving intelligence, and seamless coordination across people, agents, and services.
Yet today's messengers were designed for human-to-human communication, not human–agent collaboration. They lack the native capabilities to support intelligent, context-aware human–agent teams. It's time to reinvent messaging for the Agentic Internet.
That's why we're reinventing instant messaging at Teamily AI.
👥 People and AI agents working together in the same conversations.
🧠 Unified context and living memory across people, agents, and services.
⚡ Conversations that turn into real actions, creations, and automations, as well as agentic websites and apps as a service.
🌐 An open gateway where every agent and service is as accessible as messaging a friend.
Traditional IM connected people. AI-native IM connects people, agents, and services.
This is how we move from the Internet of Apps to the Agentic Internet.
And we're just getting started.
teamily.ai
#TeamilyAI #AgenticInternet #AINativeIM #TeamAgent #AgenticIM
Leonardo retweeted
I've seen plenty of agent demos where everything looks smooth until one agent has to pass work to another.
WorkflowArena makes those failures visible by recording every handoff, not just the polished final result.
🤝 Paid partnership
Agent harnesses now let a model spawn its own workers and write the workflow that coordinates them. But can it actually orchestrate them well?
Introducing Einsia AI's WorkflowArena, the first benchmark for autonomous multi-agent workflow orchestration on real, long-horizon work.
- 145 tasks across 8 professional domains, from SWE and Lean proofs to law and 3D
- 25 configurations of 9 frontier models in Claude Code & Codex
- Every delegation recorded: we score the deliverable and see the workflow behind it
🌐 Project: lab.einsia.ai/workflowarena
💻 GitHub: github.com/Einsia/WorkflowAr…
📄 Paper: arxiv.org/pdf/2611.12856
⬇️ Read the following threads to go through the key findings
Some multi-agent systems can get every individual task right and still fail at the bigger picture.
That's why WorkflowArena is interesting. It shows whether all those separate pieces actually come together.
🤝 Paid partnership
Agent harnesses now let a model spawn its own workers and write the workflow that coordinates them. But can it actually orchestrate them well?
Introducing Einsia AI's WorkflowArena, the first benchmark for autonomous multi-agent workflow orchestration on real, long-horizon work.
- 145 tasks across 8 professional domains, from SWE and Lean proofs to law and 3D
- 25 configurations of 9 frontier models in Claude Code & Codex
- Every delegation recorded: we score the deliverable and see the workflow behind it
🌐 Project: lab.einsia.ai/workflowarena
💻 GitHub: github.com/Einsia/WorkflowAr…
📄 Paper: arxiv.org/pdf/2611.12856
⬇️ Read the following threads to go through the key findings