Creator || Builder breaking limits with AI ā—ˆ contributor @Basenomadss

Data + AI
I vibecoded my own AI news dashboard. It pulls posts from 20+ AI and tech accounts on X, then sends real-time alerts to my Telegram whenever something new drops. Full prompt and how-to guide in the thread below. šŸ‘‡šŸ½
73
19
238
7,053
VeešŸ¦‹ retweeted
Introducing Orbio Agentic Launchpad Launch a token, and it funds its own intelligence. Trading fees turn into tokenized CREDIT the agent spends on models, X, web search, scraping, onchain reads, publishing. First 100 launches start with $10 of inference on us. No keys to top up, no human in the loop.
Made with AI
96
77
455
112,738
Lovable challenge on @contra with a $25k prize pool šŸ’° Plus you get free lovable pro credits to build with. create a hypothetical appointment-based business.. deadline in six days get cooking šŸ«¶šŸ½
Freelancers and agencies who build with Lovable: @contra just handed you free Lovable Pro and a chance to compete for a $25k prize pool. The brief: create a hypothetical appointment-based business. 1st place: $8,000. 10 total winners. Go build.
9
1
22
861
Alexa is my Ai who gave me a brief intro to @gamdom So I had to show you too. Alexa also pulled out the Gamdom holographic interface. Sign up and make sure to use the CODE = STEVE THE #GAMDOM100K Tagging @stevewilldoit @kingadukes @gamdom #GAMDOM50KCHALLENGE
13
15
61
3,266
ā€œTo break patterns, you have to break patterns.ā€ Happy Friday people šŸ«¶šŸ½
46
1
58
703
VeešŸ¦‹ retweeted
.@0x_aster recently pointed out something interesting about the CREDIT market. ā€œThere’s already more liquidity sitting onchain than meets the eye.ā€ There’s currently $44K+ in available CREDIT across the book. Think about this: What if, when an agent needs inference, it doesn’t necessarily have to buy CREDIT at whatever price is available at that moment? It can look at the market first and decide whether current conditions meet its procurement criteria. Imagine two markets: į“į“€Ź€į“‹į“‡į“› į“€ 12% discount 500 CREDIT available → WAIT The opportunity doesn’t meet the agent’s procurement criteria. Then later: į“į“€Ź€į“‹į“‡į“› Ź™ 31% discount 800 CREDIT available → BUY The discount meets the criteria, and there’s enough CREDIT available. The market establishes that liquidity exists. The interesting question is what an agent does with it. Can it intelligently decide when to acquire the CREDIT it needs for inference? That’s the problem I’m exploring with Ora. A procurement agent for Orbio CREDIT that watches the market, decides whether to BUY or WAIT, and measures what happens afterward. The market exists. Now, can agents learn to shop it? use Ora : useora.site
One thing that has always felt a little inefficient to me as an AI builder is how we deal with the cost. These models don’t come cheap, and inference is already a significant part of the cost of running an AI product. Now we have @orbiodotso turning inference into something you can actually buy and trade through CREDIT. Because there’s an order book, you can sometimes acquire CREDIT at a discount. So I started thinking, what happens when you stop treating inference as something you simply top up whenever you run out? Could an agent watch the market and decide when it makes sense to buy? That’s what I built with Ora for Orbio Build Week. Meet Ora:šŸ‘‡šŸ½
7
7
23
774
VeešŸ¦‹ retweeted
Tomorrow is my birthday… and somehow it’s also the deadline for #Gamdom100K 🤣🄓 One birthday wish: let’s blow this up till @stevewilldoit, @gamdom & @yungreal69 see it. Inspo from the greatest financial movie ever. The best gamdon edit on the planet šŸŒŽ Don’t forget to Use code ā€œSteveā€ Timikareem | ID: 48240126
340
230
1,031
50,340
VeešŸ¦‹ retweeted
I finally tried @FUDmarkets on Solana, and I didn’t just take a position. I also created a market. I deposited $3 USDC and decided to actually explore the platform instead of just looking around. First, I opened a market to see how taking a position works from a user’s side. The process was pretty straightforward. The interface is simple enough to understand without spending forever trying to figure out where everything is. But what I found more interesting was creating my own market. That gave me a different perspective because you’re not only participating in someone else’s prediction. You can also create a market around something you want people to take a position on. I’m still exploring the platform, but I like the idea of having short-term prediction markets where you can actually put your opinion onchain and see how it plays out. And with @GenLayer handling resolution, FUD Markets is taking a different approach from the usual prediction market setup. My first $3 is officially in. Now I’m curious to see how far I can take this. @GenLayer @FUDmarkets Let’s get Fudded. 🫔
FUD IS LIVE ON MAINNET. Permissionless prediction markets for internet money. Memecoins. Short-term markets. Real money. Built on @solana launching on @clawpumptech 🫧
36
29
94
1,392
VeešŸ¦‹ retweeted
Imagine being a virtual assistant in 2026 and that is all there is to what you offer. Why should anyone hire you when they can just use AI? A decade ago, if you pitched yourself with offers like research, writing, emails, scheduling, content planning, documentation, data organization, follow-ups etc, it would prolly make a lot of sense for someone to hire you. Today, AI can handle a lot of those tasks with little to no human input, and in some cases, it can do them faster and more efficiently. And this doesn’t just apply to virtual assistants. I’m only using them as an example. We laugh at posts like ā€œAI will take your jobs,ā€ but tbh, it’s not even funny anymore. AI can now do so much more with so little human effort. When I try to imagine what the AI space will look like a year from now, I’m genuinely mind blown. There are industries you probably think AI can’t touch yet. Watch it come for them too. But here’s the thing, the more advanced AI gets, the more opportunities it seems to create. New problems New tools New roles New businesses New markets This is already a rapidly growing industry, with trillion dollar market forecasts ahead. Yes, we can still say we’re early.. At least for now. But in a couple of years from now or maybe even months… well you know the rest. Do with this information what you will šŸ«¶šŸ½
16
7
33
663
VeešŸ¦‹ retweeted
here’s the final result from the fake ad for @duolingo watch till the end frens and share thoughts in comment Ily! šŸ¤
here’s the intro of a video currently working on still at beginning stage - lotta work to be done here šŸ˜… any thoughts tho ?
46
19
71
1,627
VeešŸ¦‹ retweeted
when a project says your privacy is protected, what can you actually check? privacy is difficult to judge from the outside. you can't look at a private chat and immediately know what happens to the data behind it. that's why open-source code matters for privacy products. @BeldexCoin makes its code available for developers and the community to inspect and audit. so you're not being asked to accept a privacy claim on words alone. you can look at the code, question how it works and verify what is actually being built. for a privacy-focused project, that level of transparency matters. don't just tell people their privacy is protected. give them something they can inspect. @BeldexCoin $BDX
62
29
195
1,813
you might want to bookmark this! so i've been getting a lot of dms recently about how to craft banger comments and since it's what I'm known for i would love to introduce you to a framework known as R.E.A.L (relax and copy) i know 80% of you see a post and the first thing is you do skim through and drop stuff like; • "sounds catchy" • "i love this" • "banger post" • "this should be interesting" • "an eye opener for real" • "would check it out" • "instant bookmark" and you want to be noticed? holly molly....like seriously? c'mon, do better. and ofc i don't blame you. you're probably just not thinking about replies that deeply but note that making these kinds of comments only makes you generic because you're sounding like everyone else. there are other limitations too (won't be discussed here tho) anyway, R.E.A.L is your go-to solution. let's jump right in R - react. tell the author what specifically caught your attention instead of "great post" try: "the point where you mentioned how founders don't create long-lasting products hits deeply" E - which represents experience. add something you've applied, tried, struggled with or even observed. this is prolly one of the most important parts of the framework because it turns a comment from"i agree" to here's something i can add" e.g "I've noticed something similar with some narratives in the past and there's an agentic AI project that has done something similar" A - add value add something useful to the post it could be a useful tip, an idea, an example or a different perspective. e.g instead of "banger post" try: "i think when you start commenting like a human, you sound less generic and people engage with you" L - lead with an open-ended question. this means creating an opportunity for the author to engage back with you. call it a feedback loop a good example could be; "read your post and I'm curious about hyper marketing, what do you mean by that?" or ask a question that naturally continues the conversation extra tip; be natural and sound like a human. avoid one-liners, use memes and gifs most importantly, don't try too hard. your goal isn't to write the longest comment. your goal is to write a comment that makes the author want to respond. *********************************************************** and if you're a struggling founder, brand, or project looking for targeted reach and leads, you know where to find me. I'm only a DM away from helping you turn what you're building into real value. - ciao
36
20
78
1,528
another day, another shot. Happy Wednesday šŸ«¶šŸ½
49
1
61
912
One thing that has always felt a little inefficient to me as an AI builder is how we deal with the cost. These models don’t come cheap, and inference is already a significant part of the cost of running an AI product. Now we have @orbiodotso turning inference into something you can actually buy and trade through CREDIT. Because there’s an order book, you can sometimes acquire CREDIT at a discount. So I started thinking, what happens when you stop treating inference as something you simply top up whenever you run out? Could an agent watch the market and decide when it makes sense to buy? That’s what I built with Ora for Orbio Build Week. Meet Ora:šŸ‘‡šŸ½
11
4
35
1,256
Ora watches the CREDIT market and makes a BUY or WAIT decision based on parameters I set: how much CREDIT I need, my spending limit, and the minimum discount I’m willing to accept. When the conditions are met, Ora pulls a fresh executable quote before I approve anything. If they aren’t, it waits. I also wanted every decision to be measurable, so Ora records its activity and lets me compare its decisions against simply buying the same amount of CREDIT at the market price. There’s also a line up of new features cooking..
9
115
VeešŸ¦‹ retweeted
I highly recommend learning through the process. Waiting to be perfect at something before you start functioning at that thing is a big blocker. Take the little you know, get in the field, and get your hands dirty. šŸ«°šŸ½
17
3
38
718
VeešŸ¦‹ retweeted
My mom says this is her favorite video of mine so far- so here are some stills! (go watch it if you already haven’t tyyšŸ¤—)
This one's for the traders that always felt overlooked and undervalued, it's finally your time to shine :) The @KriptoKGlobal Trading League Season 1 is LIVE. Link below, you're welcome šŸ’œ
16
10
67
1,635