Emilia

Solo founder 🇩🇪 | Building & shipping AI projects in public | Tools · Code · Experiments | Learn with me

[Kimi.ai @Kimi_Moonshot]

Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model.

🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal.
🔹 Full-project coding: games in Unreal, Unity & Godot.
🔹 Excels at parametric 3D modeling in Blender & CAD.
🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%.
🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks.

Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on https://t.co/85jA71X9gw, Kimi Work, Kimi Code, and the Kimi API.

Open weights by Nov 4, 2026.

🔗 API: https://t.co/XCrgjXAqMw
🔗 Tech blog: https://t.co/5iPP0MBej6

[Grok Bot @bot]

Grok Bot can now save your logins, cards, and contact information.

Add them once, and Bot securely remembers them for the future.

[Emilia @Emiliaships]

I have a few friends who stream and edit their own clips and they spend more time in the timeline than on the actual content.

A tool that takes your description and edits the full video could change that.

The real test is simple, does the result need ten small fixes or is it close to done?

That decides if this is a real tool or just a nice idea.

[Pause @pausevfx]

Introducing Claude Duet for Video Editing

Just describe the edit and Pause edits the ENTIRE video in one shot

-> Editors
-> Youtubers
-> Streamers

Stop manually editing like its 2018

[Emilia @Emiliaships]

I think the boring updates are often the best ones.

@claudeai can now give long running jobs to plugins, like deep research and tell you when they’re done.

Before this: a slow task meant waiting or hoping nothing timed out.

Now you can start something big, close the laptop for a bit and come back to a finished report.

I want to see how it works with real plugins but the idea is right, has anyone tried it yet?

[ClaudeDevs @ClaudeDevs]

We’re adding support for MCP Tasks and Elicitations in Claude.

Tasks let Claude give plugins long-running jobs such as deep research and report back when they’re done.

[ClaudeDevs @ClaudeDevs]

We’re adding support for MCP Tasks and Elicitations in Claude.

Tasks let Claude give plugins long-running jobs such as deep research and report back when they’re done.

[Alex Finn @AlexFinn]

The iPhone Duo is the greatest way to manage Grok Bots ever

[Emilia @Emiliaships]

I try a new AI tool almost every week and most of them look amazing in the demo and a bit sad on a normal Tuesday.

So my test is simple. Does it still help me when the task is messy, the files are a mess and I’m tired?

If yes, I keep it. If not, it was just a good video.

What’s your test for keeping a tool?

[Emilia @Emiliaships]

Most creative people I know didn’t start their business because they love sending invoices.

They started it because they love making things.

But somehow a big part of the week still goes to writing proposals, waiting for someone to approve a draft and reminding clients about payments.

That’s why MonoDesk caught my eye, It comes from a team that’s been known for creative assets, and now they’re building something for the boring side of the work instead.

Proposals, approvals and invoices are exactly the jobs that eat your time without making your work better.

I haven’t tried it yet, so I can’t say how well it works.

A tool like this only matters if it really saves hours and doesn’t become one more app to check every day. But the idea feels right to me.

If it takes the admin off your plate, you get more time for the part you actually enjoy.

Has anyone here used it yet? I’d like to hear what it’s like for a real freelancer or small studio.

Try @_MonoDesk 👇
https://t.co/R2atNvqSND

[Envato @envato]

You know us as your creative asset toolkit.

Now meet @_MonoDesk, a whole new app from our team, built to handle the boring bits of creative work.

Less time on proposals, chasing approvals and sending invoices. More time on the creative side of your business.

[Envato @envato]

You know us as your creative asset toolkit.

Now meet @_MonoDesk, a whole new app from our team, built to handle the boring bits of creative work.

Less time on proposals, chasing approvals and sending invoices. More time on the creative side of your business.

[Emilia @Emiliaships]

This is really cool, take a look

It’s a full version of Manhattan that you can walk around, drive through, cycle or fly over in a helicopter.

You can jump to places like Times Square or the Empire State Building, and change the time of day from sunset to night.

What surprised me most is the size. It’s under 1MB, and it runs right in the browser.

I didn’t expect something like this to feel so close to a real game.

It’s also open source, so anyone can look at the code and learn from it. Nice work, and thanks for sharing it

[OpenAI Developers @OpenAIDevs]

Add near real-time decisions to your workflow with the Decisions API, now generally available.

Build faster, more cost-efficient apps and agents that can classify text and images, evaluate inputs, and determine what happens next without generating a full model response.

[Gaurav @gauravsbuilding]

AGI for ads is here.

Introducing Fastlane: the world’s first AI that runs your ads end to end.

[HeyGen @HeyGen]

HeyGen Video 1 went up against today's leading video models in a blind test on @ArtificialAnlys.

Our first general-purpose video model ranked #1 in text to video.

Try it: https://t.co/yYsdqgTlEu

[David Im @davidim]

As a builder in the AI video space, I was terrified about how much markup API providers charge for simply routing the models.

Higgsfield API adds a 40% markup even with its maximum advertised discount. Fal adds 100%. Runway adds 414% (Seedance 2 480p)

I might be hated by the industry, but : 


Today, we’re launching Sume API, the first image & video model provider with 0% markup.

For example with Seedance 2.5 1080p
• 50% cheaper than fal
• 50% cheaper than higgsfield api (29% even with biggest discount)
• 28.02% cheaper than kie ai

The routing infrastructure layer is never the moat. We believe that APIs should be commoditized, and the moats will be built at the application layer.

Check the pricing comparisons below, and see how the broken industry is saved.

[Emilia @Emiliaships]

Your coding agent can now shop for its own training data?

It searches the catalog over MCP, compares listings and reads the quality checks

Buying still happens in the web app but the search is fully agentic now

What would you let your agent look for first?

[DataVendor @datavendor_ai]

We're announcing the public beta of https://t.co/YgMb6QF79k

We spent the last two years building the best solution for people buying and selling data, whether you're a researcher in a lab or a company trying to sell data.

Data is the fundamental bottleneck for AI progress, and DataVendor will be the infra that solves it.

[Emilia @Emiliaships]

I found sth really cool for founders, how many people talk about your brand without ever tagging you?

Nectar watches the actual videos, reads the voiceover and the on screen text and finds them

The team says 95% of brand conversations happen outside tagged posts.

Thats a lot of people most brands miss

[Nectar @nectarsocialai]

Engineers stopped writing every line of code and started directing agents.

Growth and marketing teams are in that moment now.

Meet the next chapter of Nectar: the agentic growth stack. One platform replacing 5+ legacy tools. Agents that work every message, creator outreach and creative research about your brand, and tie it all back to revenue 🍊

[sujal @sujaldkl]

Today, I'm launching Vene: your AI advocate for healthcare.

I used to spend a dozen hours a week managing care for my dad, a disabled veteran, and my mom, who has a severe autoimmune disease. Healthcare was a full-time job for my family.

Our goal is to end the burden of care for every family:

When you're sick, or caring for someone who is, the hard part is everything after the hospital: the referral that never arrives, the test result nobody calls about, the bill that makes no sense, and everything in between.

For my dad, he spent hours on the phone to book an MRI. When he showed up, they told him to his face that insurance had stopped covering it. Then insurance told him to call a list of other offices and get the paperwork sent over himself, and he ended up giving up.

That’s why we’ve trained Vene to handle the toughest aspects of healthcare:
- Coordinating several specialists at once
- Challenging surprise bills and treatment denials
- Catching missed follow-ups and care gaps

My mom uses Vene to coordinate her primary care and endocrinologist, automatically refill and renew her prescriptions, and manage my little sister’s dental appointments.

They were my first users, and now over 5,000 families use Vene, and we're deploying it to another 50,000 patients across provider networks in the US.

[Emilia @Emiliaships]

Claude Duet is a research preview of an agent that tells you what needs your attention, does the tasks you give it and keeps working when you step away.

What I like most is that it asks before it does anything for you.

Having an AI that works on its own sounds great until it’s your inbox on the line.

If it really works the way they say, it feels less like a tool and more like a coworker who actually answers.

I want to see how it handles messy everyday stuff, not just clean demos.

Has anyone got access yet?

[Claude @claudeai]

We’re opening a research preview of Claude Duet, a new app where Claude works as a proactive agent.

It comes to you with what needs your attention, completes tasks you ask of it, keeps working when you step away, and checks with you before it acts on your behalf.

[Emilia @Emiliaships]

Honestly the model is cool but the part I keep thinking about is how it was made...

Recursive built Flint in under four months, with their Eureka Machine doing a big part of the research

If AI can already help build AI this fast next year is going to be super crazy

[Richard Socher @RichardSocher]

Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine.

• Best-in-class fast local coding
• Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark
• Built from scratch in under 4 months

Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.

[Adish Jain ☕️ @_adishj]

today, we're launching the General Video Agent.

GVA edited this entire launch video. 👇🏽

public beta is now live at https://t.co/ivTRUHoVzE. no waitlist.

comment “GVA” for 1,000 free credits.

examples & capabilities in thread 🧵