slavique

compare yourself only to who you were yesterday — not to others

@FarzaTV brilliant

I'm blown away at what ppl are using this for!! I built it as a learning tool. But people seem to really love using it as an AI interface that isn't chat that can work in their program of choice. Examples of usage so far: - A Mom building her first app on Lovable - A dentist debugging his OpenClaw setup - A photographer getting feedback in Lightroom - A person learning to animate SVGs in Framer - Founders keeping track of their todos. - Designers getting feedback in Figma - A student outlining her thesis in G-Docs - Traders analyzing live stock charts And A LOT of people using it to advise them on how to best reply to messages in Slack/Email. Super cool. The people yearn for a non-chat interface haha. Also, it's kinda crazy how as the founder you really don't know what the product is until you put it in the hands of users. The minute it's in the hands of others, it's theirs now! And that's really where you find out what it is.

@bronfirstring love the idea of sparks

this knowledge base narrative being everywhere in my timeline last few days got inspired to share my vision on how you can define “ideas” so that it makes sense for LLM to work with. it was formed in last couple of months when i was working on the pet project trying to fix my attention so what’s my point we are witnessing how knowledge base ops are on its way to be pretty much solved, LLMs are capable of it, they are already good enough - summarizing incoming content, organizing structure of files and making it connected by wikilinks and it seems to be just on time. it's just too much content today. Andrej's tweet has 24k likes and 44k bookmarks. people save more content than they are actually capable to digest but how do we understand content is valuable enough to save? i think it’s about the moment when it sparks your curiosity. that’s real human curation. personal relevance do matter. and it vanishes the moment you hit save this tweet you lose this spark when you return to this content (if you actually do). then you end up having a whole rawContentLink squad rotting in your tg saved messages, giving you absolutely no meaning a raw link can provide LLM with content. but not with "why you saved it", "what project in your head it connects to" or "what you intended to do with it". that context only exists in the moment of capture. and no format currently captures it— it’s just another link, video, article, notes, transcripts. content type doesn’t matter. it’s still a low-signal resource for an LLM, just content digestion and retrieval. to address this i've created a structured agent-digestible format of .md doc that not only makes sense to LLMs, but for human beings. so i frame it as spark. spark - is a unit of interest. github - github.com/skyhook19/spar… it's still early, so i'll be grateful for any suggestions and open-source impact spark.md is an open format for people to be able to easily manage their interests with any modern tools spark helps you to answers three questions: - what is it? - why does it click? - so what? (actionable) that's the difference between just a link and a unit of thought. spark won't rot in the saved messages, because now we have actually something to do about it. as your base grows, sparks naturally tend to unite and transform into a higher-level entities - projects, habits, research. and they are united not by structure, but by meaning i came up with this idea when i was brainstorming a name for my pet project catchsparks.app which i hope to finally finish in next two weeks it's like a smart inbox app, which can catch everything that sparks you, and help you curate this as a content collection. and it's open - you can export and delete your data anytime, in open format that makes sense both to humans and agents tbh i don't know if this really a thing, but maybe you've noticed this too - you feel like you need to save all the content, but actually never consume it. this likes/bookmark ratio speaks for itself spark concept addresses that moment. i think that by understanding "what caught your attention", you connect "the content" and "why it actually matters to you". that context is the real signal, and it's the one we lose first i think that's worth preserving

@Lovable see u in paris

Introducing the Founder Series: a new set of global events, by founders, for founders. You’ll hear from people who have already turned their product into a real business, and then get the chance to build your own. Find a Founder Series event near you: founder-series.lovable.app

Say gm to our ETHGlobal Cannes 2026 finalists 👋

a moment to celebrate and remember. now back to building

got to finals with trymaki.xyz super grateful 🙏 thanks @ETHGlobal and the team, what an event!

@kidrulit great read, thanks for putting this together

@binji_x @ETHGlobal 🙏

Always an honor to be a judge here. The projects I judged were ENShell & Maki. Both of these were focused on Agentic privacy and security on Ethereum. I have often said openclaw is one of the best things to have happened to crypto, since it makes the need for privacy and security so apparent. Great seeing people build on that.

With self-serve access to ChatGPT Ads coming in April, the PPC game is going to change forever It took our dev team less than 36 hours to build Ads Analytics once we got access! These guys ship 🦾

@rmrf_404 @bcherny lol

we've already heard from @bcherny that Claude writes himself. Does this mean Claude is lonely and decided to get himself a Tamagotchi system called "Buddy"? :D

bro got bored coding in silence

@rmrf_404 🙏

@serafimcloud @ExaAILabs great use case

We had no idea when engineers from OpenAI or Stripe signed up. So we built a lead research agent. Claude Code + @ExaAILabs researches each signup by email and pings Slack when someone interesting appears. Template on 21st 👇

@rmrf_404 i hate this part

AI makes building faster. It does not automatically make thinking better. That part is still on you.

@SamiraBehrouzan @speedrun 🫡

if you can reply to this message you can apply to @speedrun.

@Kitty_Mayo_ @ncsh let’s go ☀️🌊

80 engineers. Teams of three. Five tracks across robotics and AI. Travel, accommodation, food and compute all covered. projectbarcelona.co

the real battle in tech isn’t ai vs humans. it’s human agency vs systems that quietly take it away.

@mlfrg wow, super fun!

introducing Ficus: vibe code multiplayer apps to interact with your audience. Give people a QR code, they connect with their phones or laptops during a meeting/presentation and can collaborate in real-time. You can build polls, multiplayer platformer games, ice breakers, quizzes, collaborative drums machines, oiia cat multiplayers, anything real-time

today i have nothing to say

@marclou now build a cli x client so people think ur coding

I built this stock market app so I can open my web analytics in public without people thinking I’m unemployed.