Marko Srsan

32.2K posts

Marko Srsan banner
Marko Srsan

Marko Srsan

@msrsan

COO @memgraphdb, ex-director @Techstars, founder @TimePrepCFA, drummer and a passionate FC Bayern fan :)

London, UK Katılım Haziran 2010
1.9K Takip Edilen2.9K Takipçiler
Marko Srsan
Marko Srsan@msrsan·
@macrofactorapp⁩ - this is what I get when I want to use the AI food recognition. The feature worked fine until yesterday but today it just goes blank/white whenever I want to take a photo. Can you look into it, please?
Marko Srsan tweet media
English
0
0
1
51
Marko Srsan retweetledi
Matija Sosic
Matija Sosic@MatijaSosic·
Can AI create a sleek, $25k-design-agency website vs the usual “hairdresser salon” slop? I tried five AI tools. It was impressive and nerve-wracking at the same time. Here’s a breakdown:
English
1
12
34
3.9K
Marko Srsan
Marko Srsan@msrsan·
Who's got a SORA invite code? 🙏
English
1
0
0
130
Marko Srsan
Marko Srsan@msrsan·
@taykv2 Hey Kev - in light of the most recent Kuzu news, let me know if you looking for a performant graph DB as an alternative. I'm with Memgraph and happy to help.
English
0
0
0
41
kev
kev@taykv2·
Q: Why Neo4j? A: I had some familiarity with it, but we switched to Kuzu (@kuzudb) to improve performance
English
5
0
28
2.6K
kev
kev@taykv2·
Some quick context to the viral video and answers to frequent questions: When that video was taken 1.5 months ago, I had been working on it as a fun project for 1 week (‼️). My v1 is significantly more fleshed out and I now have a team of 4 working on a full rewrite.
kev tweet media
English
16
16
289
42.1K
Marko Srsan retweetledi
Matija Sosic
Matija Sosic@MatijaSosic·
We’ve run 11 Launch Weeks at Wasp - probably more than anyone except @Supabase. From the outside, it's all fancy tweets, memes and PH launches. But here’s what it actually feels like to run 11 of them, and why its valuable even if only five people join:
Matija Sosic tweet media
English
6
16
30
5.3K
Marko Srsan retweetledi
Matija Sosic
Matija Sosic@MatijaSosic·
This is why we're building wasp.sh - a new, truly full-stack framework. Devs want Rails/Laravel experience for JS, but all the attempts so far have failed b/c either the timing was wrong, or the core team gave up (lost motivation, no money). We've been pushing for almost five years and raised >$5M to have serious folks work FT on it. We're getting there, commit by commit.
DHH@dhh

People dismiss so many good ideas with "we tried that, didn't work", only to be proven wrong by someone with enough determination and the right angle. Turns out we DID need another Linux distro 😄🚀

English
20
27
134
42.1K
Marko Srsan
Marko Srsan@msrsan·
Great keynote by @swyx sharing the State of the Union (AI) as the year of agents is upon us. Day 2 today! Let's goooo! #AIEngineer #AIEParis
Marko Srsan tweet media
English
0
0
0
42
Marko Srsan retweetledi
Brain Aboze
Brain Aboze@AbozeBrain·
Thrilled to be at @aiDotEngineer (AIE) Paris 🇫🇷, the first AIE edition outside the US 🇺🇸, hosted at STATION F and organised with Koyeb. It was inspiring to hear the origin story from AI Engineer co-founder Benjamin Dunphy. His vision of creating a global movement, not just an event, has clearly come to life. Partnering with Koyeb, an AI infrastructure company and one of Station F's top startups, was a brilliant move to ensure a high-signal, high-quality experience. As Yann Leger, @gokoyeb’s CEO, highlighted, with 70% of attendees coming from outside France, this truly feels like a global gathering. Also, a welcome note came from Marwan Elfitesse, Head of Startup Programs at @joinstationf. He perfectly set the stage, reminding us that STATION F is a major AI hub, being the birthplace of alumni like @huggingface and home to the event's key organisers, Koyeb. The keynote from @swyx​ (Shawn Wang), AIE co-founder, on the state of AI Engineering was a fantastic deep dive. He painted a clear picture of the future: we're in the midst of an "epic infrastructure buildout," and the next decade will be defined by building AI agents with memory, planning, trust, and orchestration. His breakdown of the emerging "LMOS" (LLM Operating System) and the open debates around evals vs. vibes, context engineering, and the need for fast agents gave us so much to think about. Rounding out the evening, Lélio Renard Lavaud from @MistralAI AI offered a powerful perspective on enterprise adoption. He emphasised that overcoming challenges like data silos and observability is key, as well as governance, speed, and education. His talk was a great reminder that success isn't just about powerful models, but about building a robust "flywheel" to build, deploy, observe, and improve AI systems in the real world. On a personal note, I met @marlene_zw again, the last time was #Pycon Nigeria in 2019. One of the earliest pioneers of African AI communities, and still inspiring today. It was wonderful reconnecting with her and so many other brilliant people here. Huge congrats to the AI Engineer and Koyeb teams for bringing this community together. Looking forward to Day 2! If you’re here, say hi! I’d love to swap notes 😃 #AIEngineer #AIEParis #LLM #StationF #Koyeb #MistralAI #AI #Agents #AIInfrastructure #Community
Brain Aboze tweet mediaBrain Aboze tweet mediaBrain Aboze tweet mediaBrain Aboze tweet media
English
1
5
29
4.5K
Marko Srsan retweetledi
Sam Altman
Sam Altman@sama·
gpt-oss is a big deal; it is a state-of-the-art open-weights reasoning model, with strong real-world performance comparable to o4-mini, that you can run locally on your own computer (or phone with the smaller size). We believe this is the best and most usable open model in the world. We're excited to make this model, the result of billions of dollars of research, available to the world to get AI into the hands of the most people possible. We believe far more good than bad will come from it; for example, gpt-oss-120b performs about as well as o3 on challenging health issues. We have worked hard to mitigate the most serious safety issues, especially around biosecurity. gpt-oss models perform comparably to our frontier models on internal safety benchmarks. We believe in individual empowerment. Although we believe most people will want to use a convenient service like ChatGPT, people should be able to directly control and modify their own AI when they need to, and the privacy benefits are obvious. As part of this, we are quite hopeful that this release will enable new kinds of research and the creation of new kinds of products. We expect a meaningful uptick in the rate of innovation in our field, and for many more people to do important work than were able to before. OpenAI’s mission is to ensure AGI that benefits all of humanity. To that end, we are excited for the world to be building on an open AI stack created in the United States, based on democratic values, available for free to all and for wide benefit.
English
1.3K
2.1K
21.4K
2.7M
Marko Srsan retweetledi
Bearly AI
Bearly AI@bearlyai·
insane day for AI releases
Bearly AI tweet media
English
22
14
89
46.5K
Marko Srsan retweetledi
17 years of song a day
17 years of song a day@songadaymann·
i vibe coded a little game called Coldplay Canoodlers you're the camera operator and you have to find the CEO and HR lady canoodling 10 points every time you find them 👇link
English
627
4.1K
58.5K
5.1M
Marko Srsan retweetledi
Dave
Dave@GamewithDave·
Are you old enough to remember when GTA looked like this?
Dave tweet media
English
2.3K
1.4K
45K
2.2M
Marko Srsan retweetledi
Jack Bridger
Jack Bridger@jacksbridger·
Quick update: @ittybithq are acquiring StreamPot and StreamPot will be closing down at the end of October. Thanks to everyone who used StreamPot, tried it out, gave me advice, contributed etc. I won't pretend this is a big success but I've known @PaulAntWilliams since day one and I'm happy ittybit are providing a strong exit option for StreamPot users (alongside self hosting the open source version on railway/northflank etc.), and I'm happy to have a team I can cheer for and refer people to because I still think it's a huge problem that isn't solved yet. I recorded a scaling devtools episode with Paul where I share the whole story - will release soon. But tl;dr: - what went right: found a big annoying problem for people and built a simple to use tool - what went wrong: not the right team setup to build a stable video infrastructure tool There were ups and downs and late nights resetting servers but overall it was one of the funnest things I've ever done and I couldn't be more grateful for all the memories and lessons. And.. StreamPot has also given me a lot more questions and empathy for guests on scaling devtools.
English
18
4
58
4.9K