X Stats

Login Sign Up
@tonyennis

Tony Ennis

@tonyennis

4,360 Tweets
2,743 Following
2,723 Followers

Building software teams & mobile apps Prev: Built a bank in Asia YC S21

Filtered by topic: AI & Future Tech ✕ Clear filter
Date ↓ Tweet ❤️ 🔄 💬 👁 📝
Dec 12 But I thought you *needed* to write everything in typescript to not get left behind by AI assisted coding? 2 0 0 404 106 .
Nov 26 Claude being very poetic this morning https://t.co/Zs47OzKZtZ 2 0 0 270 61 .
Nov 08 Seeing this "AI written smell" everywhere now. Just me? https://t.co/xdGloqHoQK 2 0 1 246 79 .
Nov 01 Man AI has gotten crazy good 0 0 0 349 28 .
Oct 23 The new pattern of measuring AI progress by how many minutes it can work consecutively for reminds me of… https://t.co/LYyM4iurpp 1 0 0 170 129 .
Oct 17 Another few hours trying new AI tools today - this time the AI avatar tools - every single one littered with bugs in the onboarding. The enshittification continues... https://t.co/Faxp9CuN27 0 0 0 390 192 .
Oct 15 Isn’t it kind of bad that the AI czar is openly conspiratorial against the second most important AI company in the world and the US? Shouldn’t he at the very least be engaging in dialogue? 1 0 0 412 188 .
Sep 20 @raagulanpathy Non crypto people don’t understand “it’s high risk because it’s a way to pull cash easily” - they just know their bank works fine but Kast doesn’t. I’m also confused what this means in this context - is it money laundering or tax evasion or something you’re trying to protect… 0 0 0 0 291 .
Sep 14 Another weekend where I've been both completely blown away (@interaction) and extremely frustrated by new AI stuff https://t.co/b9TF92uBAa 0 0 1 458 138 .
Aug 27 One argument for why AI customer support tools are considered so good, is that the bar for *human* customer support is also extremely low. https://t.co/uD9OeLBK5G 3 0 2 342 162 .
Aug 21 Friend told me I had some tax benefits I didn’t know about. Used Claude to verify step by step and provide links. Asked my Irish tax advisor to process some paperwork. They scheduled two calls and on the second one laughed at me for using ChatGPT and rubbished the idea. “Was your friend drunk when he told you that. You shouldn’t use ChatGPT… We won’t be out of a job yet” Just got confirmation from my Dutch advisor that my friend (and Claude) was in fact correct. Pro tip if you’re an “expert” particularly if you sell advice/hours: Don’t do this 2 0 0 344 556 .
Aug 03 I don’t understand. Didn’t one of the js influencers say recently that AI doesn’t work if you don’t have typescript? 2 0 1 343 116 .
Jul 14 It feels like The Boiling Frog effect should be much more mainstream - potentially taught in schools. It's very hard to discuss slippery slope scenarios with people who are unfamiliar with the concept, and seems there are going to be more and more of them with AI. 0 0 0 230 264 .
Jun 16 Have a hunch @-mentioning will become a huge part of consumer AI User Interfaces - it's a much more seamless flow for including exactly the context you want "Hey ChatGPT, can you message @mark the takeaways of @notes-jun21 and book dinner with @amy using @dinner-preferences" 2 0 0 209 277 .
Jun 08 Wrong takes I see often - AI in its current form isn’t very useful - Large Language Models are the path to AGI Both are false, and believing the inverse of both is also perfectly acceptable 1 0 0 458 192 .
Jun 05 It's a shame that crypto earned itself such a terrible reputation - I'm not a crypto fanboi by any means, but I do think this use case (usdc transfers for ~free) is still being slept on because people think crypto=bad 2 0 2 345 217 .
Jun 04 Might fuck around and have Claude write this research as a tweet storm 0 0 0 471 70 .
May 30 Using AI for non trivial long running work follows a similar graph to the Gartner hype cycle But you have to actually use it for hours a day, for a sustained period to internalise that. This might explain why CEOs of larger companies (including the AI cos themselves) are so hyperbolic - they’re not using it first hand for long enough to see it’s imperfections, and they’re drowning in the noise from all the people who are atop the peak of inflated expectations, and assuming they’re at the bottom of a hockey stick. 3 0 0 192 520 .
May 26 @aidenybai You forgot “Have novel thought, get sucked into creating tweet about it” 1 0 0 0 83 .
May 24 May just have invented a much faster way to interact with LLMs? Do people do this? Is it just a coincidence this worked? cc @bentossell https://t.co/djGnMPfxdh 2 0 1 384 160 .
May 19 @usescoutai @RobFarlow Study here: https://t.co/oSdmL7eiIm 0 0 0 112 58 .
May 19 In 2025, the top open source browser agent can still only reliably complete tasks 30% of the time. To replace real employees, this will need to get closer to 100%. The top closed-source is still only at 60%. This reflects our real world experiments at @usescoutai HT @RobFarlow https://t.co/8T6Yuh5KOM 1 0 1 227 302 .
May 07 I try not to be mean on here, but it’s hard to highlight obviously wrong statements like “Basically everything you see will be generated by an Llm” without coming off as that. That idea - that humans don’t care about predictability in their interfaces - defies a *lot* of patterns to date. If you think it, write a post or tweet thread that breaks down why - these offhand comments making big claims come off as extremely hubristic and naive. 0 0 0 272 442 .
May 07 😂 The @AnthropicAI system prompt basically does what I did with https://t.co/NESs4jkTw4 https://t.co/IHSo4EN499 1 0 0 207 111 .
May 02 Whatever LLMs do is not *reasoning*. Human intelligence doesn't work like this. If a human said this you would think they were having a stroke. https://t.co/0jL8wSZkHb 0 0 0 202 167 .
Mar 13 This 👇 The main thing that kept me somewhat open to the idea was seeing posts coming out of OpenAI, interviews with Sam & other AI CEOs. But after multiple iterations of clearly incremental improvements, their credibility is totally gone at this point. 0 0 0 287 257 .
Feb 28 When it comes to AI for coding, all roads lead back to Claude Sonnet 3.5 (including sonnet 3.7) 1 0 0 129 95 .
Feb 05 @perplexity_ai 🤷‍♂️ https://t.co/MZ6LaPCVs8 0 0 1 223 43 .
Nov 02 PSA: Don't use ChatGPT for estimates. https://t.co/ge9Nt4iwAZ 1 0 0 243 61 .
Oct 04 Just had a look at the PhotoAi reviews on TrustPilot. I think there are a lot of things to learn from Pieter but his focus on automate-customer-service is his biggest weakness and not IMO something to emulate/glorify. The HeadshotPro reviews show how to do it well. 0 0 0 268 267 .
Mar 19 No way this isn’t an llm trained on Naval’s stuff 4 0 0 423 49 .
Jan 23 As cool as current AI tools are, the fact that their only knowledge is what already exists means they can't imagine or create new things https://t.co/T7ZC0UdMB7 2 0 1 328 160 .
Jan 09 @Shpigford @readme .yml erb* file 🤦‍♂️ 0 0 0 0 38 .
Jun 30 @Mqsley @ryanckulp ChatGPT should solve this for you no? 0 0 0 0 56 .
Mar 31 Still thinking about this - will become more important as a decent chunk of point-and-click interfaces get enhanced/replaced by natural language with AI. 1 0 0 403 153 .
Jan 19 Drew this last week as I was complaining about the lack of thoughtful commentary on our "AI future". Lots of people in Category A. Haven't found any in Category C yet, but starting to see more in category B. This from @elidourado is great. https://t.co/2wXw66gafV https://t.co/1lyiqqwTYV 1 0 0 0 288 .
Oct 13 Prediction: There's going to be a big reckoning for many of these web3 companies in the future - particularly those who claim to be philosophically motivated, when they're forced to choose between their values (and/or users) and making their investors money. https://t.co/Ketbl67gLx 4 0 0 0 282 .
Sep 29 @andrasbacsai 💪 love it 1 0 0 0 23 .
Aug 27 @yangpten Kinda like https://t.co/RtWDWQ1PYb ? 2 0 0 0 46 .
Jul 04 @peteromallet No mainly because I know you're a gpt-3 nerd, and also the joke wasn't laugh out loud funny - and that's your brand. 7 0 0 0 130 .
Jul 04 @peteromallet GPT-3 0 0 0 0 19 .
Jun 09 @Mqsley @joincolossus @gabrielleydon You should. Although it’s reasonably web3 heavy and not sure that’s your thing. But he makes some great points around “if your innovation relies on a smooth user experience, it’s probably not innovation” 3 0 0 0 240 .
Dec 05 @heyellieday This confuses me. What constitutes making these web3-native? I used to think it was storing the underlying data on chain but we know that’s cost prohibitive. So is it something else, like using web3 payments or identity? 5 0 0 0 233 .
Nov 21 Visa has over 20,000 salaries to pay, so measuring transaction cost as % of volume moved is (somewhat) justifiable. Blockchains *don't employ people* - why go to all this effort to improve/replace the old system and set the bar at the same place as the old one - makes no sense https://t.co/H5dWi3KUM8 2 0 0 0 301 .
Nov 12 @Shpigford @VibeWarsNFT @LaserTweets Looks cool! Out of interest, how is it monetised? Couldn’t see anything on the site 0 0 0 0 120 .
Nov 01 @dabit3 @WebSummit @CccLisbon @SolanaConf @graphprotocol @theklineventure @reemisradd @martint Where might one catch you for web3 chat and/or beers between now and Fri? 1 0 0 0 168 .
Sep 27 ❤️ Love this from @daxon. One of the best examples I've seen of using NFTs & their community for good https://t.co/ydE74KjPL2 4 0 0 0 129 .
Sep 12 "Programs", aka Smart Contracts, are just functions - they take some inputs, run some logic, and return outputs. They're written in Rust and stored and run on the blockchain itself, and can communicate with each other. There are "System" programs, and "User-submitted" programs. 0 0 0 0 278 .
Jul 27 Request: An online payment flow that works like Omniauth (Sign in with X). When you sign up for a new service, you simply authorize your wallet and the new service gets the authority to debit it. No need for cards/middlemen. Presumably would only work in web3 land. 7 0 0 0 265 .
Mar 18 @callmehouck @sam_kirschner I'll just leave this here 😁 https://t.co/OIpqx0Vmir 0 0 0 0 80 .
Dec 08 @ani_pai Congrats! 2 0 0 0 18 .
Oct 12 @hhariri Probably coming at this from a different angle but... https://t.co/zIdUoY7jai 25 2 0 0 87 .
Oct 03 @stevecheney GPT-3? 0 0 0 0 19 .
Sep 14 @Goltra 😂 hadn't seen that before. Sounds like it was written by GPT-3 1 0 0 0 70 .
Feb 03 @Conaw could you elaborate a bit on this? Was this a data leak? Should I not be using Roam for private notes? From https://t.co/0Ole6Q4MmL https://t.co/4xP07p0CxC 0 0 0 0 162 .
Feb 02 😱 Mumbai did it! ^ https://t.co/Lox6ccZvN2 1 0 0 0 43 .
Dec 19 @callmevlad @Conaw @howietl Feels a bit wordy though? This’ll become a much bigger category in the next few years and we’ll need something catchier to lubricate mainstream awareness. You’ve got the authority to shape the language! Haha can we get it down to two or three syllables? 😁 2 0 0 0 283 .
Sep 17 @_SimonLau @otter_ai @SorayaRoberts Hi Simon, I don't see this option in my settings. Is it limited to the pro version? Thanks! 0 0 0 0 127 .
Apr 20 @anikasai https://t.co/vxnUYoyMgz 2 0 0 0 33 .
Apr 17 @michael_nielsen This thread between @yorl and @tage_rai is a good example https://t.co/cYO46VpROR 1 0 0 0 98 .
Jul 11 RT @BenedictEvans: People always imagine automation will give us people-shaped robots that will try to kill us, and we always end up gettin… 0 0 0 0 140 .
Mar 30 2015 ✅ Memes/gifs in place of language 2016 ✅ Popularisation of AI voice assistants 2017 ✅ Affordable personal tracker drones 2018 - Weebo? https://t.co/azG69I0m2b 1 0 0 0 164 .
Mar 30 Asking an AI "Why is this tweet funny?" would be a very good Turing test. https://t.co/qy1cYCA0TT 0 0 0 0 97 .
Feb 10 @jaynesara20 @ThousandNetwork is that @niallmcardle90 spotted in the background? 0 0 0 0 80 .
May 02 @EmiratesSupport think my wallet was left on flight just in from Dubai to Dublin. Tried calling but left on hold. What should I do? 0 0 0 0 131 .
Nov 29 @niallmccarthy89 @digo995 I know gas aren't they. Voting opens next monday so you can like it then #XMASFACTOR 0 0 0 0 110 .