OpenPixels @OpenPixelsApp

AI colourisation can help you see an old black-and-white photo differently. It cannot tell you exactly what colour every coat or wall really was. Keep the monochrome original and label the colourised copy as an interpretation.

I built TapDraft, an AI copilot for growing on X. It's made for people who build in public, and it does that one job. tapdraft.ai/?ref=x-promo

@shiyam_kashfiq @designby_rubik Creative tools should empower uniqueness, not conformity. Open-source solutions can help founders maintain their distinct identity without sacrificing control.

@htahir111 AI can enhance creativity, but it should empower users with open-source tools that prioritize control and privacy over ads or subscriptions.

A simple photo naming habit: subject, approximate date, version. For example: family-picnic_1998_original and family-picnic_1998_restored. Future you should not have to guess which file was edited.

Anthropic should pause its initial public offering until it has fully addressed risks around artificial intelligence, according to New York City’s comptroller. Ed Ludlow reports bloom.bg/4scxAVq

@adarsh1429 Identifying key trends is crucial, but don't overlook the importance of empowering users with tools that keep their data private and workflows efficient.

Working through several photos? OpenPixels supports batch processing, a ZIP of the results, and a CSV of each photo’s measurements. Processing happens on your own device. openpixels.app

I lived through the dot-com boom. But something happening in AI right now is unlike anything I've seen before. And it's not the valuations. I graduated from MIT and spent 15 years trading derivatives on Wall Street and in London. I've watched fortunes made and lost. But look at what's happening in private markets today. Last quarter, just 27 startups raised over $1 BILLION each. Together, they took roughly a THIRD of all venture capital invested worldwide. Meanwhile, over 500 venture-backed startups have been acquired by OTHER private startups this year. Companies that haven't even gone public are buying their competitors. And something fascinating is happening behind closed doors. The best founders aren't chasing investors anymore. Investors are chasing THEM. Of course, this isn't true for every startup. Most founders still struggle to raise money. But when you're building something exceptional and have five investors competing to fund you, another cheque isn't very exciting. You start asking a different question: **What can you offer me that money can't buy?** This is exactly why ambitious AI founders come to pitch our Investment Club every Monday. Through London Real TV, we've spent over a decade building an audience of millions. That's distribution a startup can't simply buy overnight. And increasingly, that's what founders want. The biggest shift in private investing isn't how much money is available. It's that the best founders can afford to be selective about whose money they take.

TxFlow team listened to the feedback regarding their initial points distribution design and it looks much better Was waiting to see what they would do before posting as I've talked with team about their first design which was not really very well balanced, especially with a tremendous weight on referral points Some key points: > We should get the early phase points landing this week. Those who used @TxFlow_L1 before will get rewarded, not sure of the total pool of points >S1 has an ending date, Q1 2027, although I don't expect TGE so soon New points distribution, with snapshot on Monday and distribution on Tuesday: > 1.25M points for traders every week (82%) > 152k ref points per week (10%) > 8% to growth reserve basket, that could be or not be fully used, unused points rolling over to the next month and not used share at the end of the season gets burned Points will be scored by a mix of volume, OI, fees paid, capital kept on the platform, consistency and referrals I've been using TxFlow mainly to DN RWA pairs with Arcus 5% fee rebates: app.txflow.com/r/TXGYO1

TxFlow @TxFlow_L1

a week ago

UPDATE: Our priority stays the same: rewarding real trading fairly. We shared the rules early to hear from you, and we're making focused updates that support that goal. Here are the main updates and clarifications addressing key questions: · Updated pool allocation with more

105 45 344 66K 35

Three words I’m tired of hearing: 1. Analytics 2. Silverfield 3. Cramsey What’s yours?!

If you could give one old photo a careful refresh, which would it be: a family portrait, a childhood snapshot, or a picture of a place that has changed? What makes that photo worth keeping?

re-audit every live deal’s revenue definition (ARR vs. run-rate vs. gross bookings) the way the market just audited OpenAI.

WWE’s first ReelShort collaboration "Agent of Shadows," starring Drew McIntyre and more WWE Superstars, premieres October 23.

అడవి తల్లి బాట 2.0! #AdaviThalliBaata

Need a transparent cut-out? Choose PNG. Sharing a photo without transparency? JPEG may be a practical option. Pick the format for the job, and keep a separate original before making smaller sharing copies.

AI was supposed to kill brand design. Now anyone can generate a brand in minutes. And that’s why so many look identical. Ambitious founders don’t want to blend in; they want out. Generic is free. Distinctive costs.

keyboard_arrow_left Previous keyboard_arrow_right Next

@iamshraey_ Local AI can really shift the paradigm—imagine taking back control of our devices while optimizing resources. The future looks decentralized and user-centric!

OpenPixels shows a before/after slider and quality measurements. PSNR and SSIM compare the result, reduced to the input size, with the input. Useful checks, but not proof that newly estimated fine detail is historically accurate. openpixels.app

DigitalOcean and @nvidia are hosting the Open Intelligence Summit today, gathering teams around open models, data, and agents. The core pitch is ownership: intelligence powering products should stay portable and controllable rather than locked into closed stacks. Tracks cover building open data infrastructure, serving specialized open models for agents, and running production agent workflows. Live streams are up for the keynote and the Build, Serve, and Run sessions.

DigitalOcean @digitalocean

6 days ago

Who gets to shape the future of AI? Today we're bringing together some of the biggest voices in open models, data, and agents to answer that question, including @rbren_dev of @OpenHandsDev, @zlxi02 of @inferact, @GerryFPatia of @boltdotnew, @vinaygo of @Amplitude_HQ and @latkins

92 107 365 2.0M 278