πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede) header
πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede) profile picture

I mostly tweet about software development and technology start-ups. Consultant @ VNTRS Stockholm

Stockholm, Sweden
Joined January 2007
Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)Photo by πŸ„°πŸ„»πŸ„΄πŸ…‡ (@AlexHofstede)
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
I built this thing called Clicky. It's an AI teacher that lives as a buddy next to your cursor. It can see your screen, talk to you, and even point at stuff, kinda like having a real teacher next to you. I've been using it the past few days to learn Davinci Resolve, 10/10.
1,303
1,163
584
15,721
3,007,103
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
Milla Jovovich (actress from The Fifth Element) created a world-beating Claude memory system with @bensig?! - 100% on LongMemEval β€” first perfect score ever recorded. Free and 100% open source. Github link in the quoted post from Ben. I'm keen to hear how it works for you.
My friend Milla Jovovich and I spent months creating an AI memory system with Claude. It just posted a perfect score on the standard benchmark - beating every product in the space, free or paid. It's called MemPalace, and it works nothing like anything else out there. Instead of sending your data to a background agent in the cloud, it mines your conversations locally and organizes them into a palace - a structured architecture with wings, halls, and rooms that mirrors how human memory actually works. Here is what that gets you: β†’ Your AI knows who you are before you type a single word - family, projects, preferences, loaded in ~120 tokens β†’ Palace architecture organizes memories by domain and type - not a flat list of facts, a navigable structure β†’ Semantic search across months of conversations finds the answer in position 1 or 2 β†’ AAAK compression fits your entire life context into 120 tokens - 30x lossless compression any LLM reads natively β†’ Contradiction detection catches wrong names, wrong pronouns, wrong ages before you ever see them The benchmarks: 100% recall on LongMemEval β€” first perfect score ever recorded. 500/500 questions. Every question type at 100%. 92.9% on ConvoMem β€” more than 2x Mem0's score. 100% on LoCoMo β€” every multi-hop reasoning category, including temporal inference which stumps most systems. No API key. No cloud. No subscription. One dependency. Runs on your machine. Your memories never leave. MIT License. 100% Open Source. github.com/milla-jovovich/me…
Readers added context they thought people might want to know
The claimed 100% LongMemEval score uses targeted fixes for the 3 failing questions and LLM reranking (held-out score: 98.4%). The 100% LoCoMo score uses top-k=50 exceeding session count with reranking (honest top-10 no rerank: 88.9%). github.com/milla-jovovich…
223
745
226
6,729
1,725,881
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
We are introducing EU Inc. To make building and growing a business across the EU faster, simpler, and smarter. πŸ”Έ Start a company in less than 48 hours πŸ”Έ No minimum capital requirement πŸ”Έ Fully online and borderless
597
1,138
549
7,342
2,354,653
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
Awwww did someone take your hard work and use it to train a model to mimic your expertise without compensation
Google has revealed that "commercially motivated" actors attempted to clone @GeminiApp by bombarding it with over 100,000 prompts. This "model extraction" attack aimed to steal the AI’s proprietary logic and reasoning capabilities, particularly in non-English languages, to train a cheaper, unauthorized copycat model. The attackers systematically mapped Gemini’s response patterns to create a synthetic dataset for fine-tuning smaller, open-source models. Google’s Threat Intelligence Group detected the coordinated activity and blocked it, labeling the incident a direct attempt at intellectual property theft. Beyond commercial cloning, Google’s report noted a rise in state-backed threats. Groups from Russia, China, Iran, and North Korea are increasingly using AI to refine phishing campaigns, perform reconnaissance, and assist in writing code for malware. Source: Ars Technica
101
4,213
28
41,734
1,026,835
Met three ppl this month complaining their AirPod mic was starting to fail. Either everyone’s are gunking up at the same rate… or Apple’s planned obsolescence timer kicked in? πŸ˜¬πŸ€”
17
UX pet peeve πŸ”₯ Relative date and time are fine for readability sometimes, but include the actual date in a title, aria and/or hover element for those times when the user needs the exact timestamp
18
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
Na het zien van indrukwekkende AI-films deze zomer, werd ik geΓ―nspireerd om 2 dagen mijn tanden te zetten in het tot leven wekken van de schilderijen van Vermeer met #RunwayGen3. Dit is het resultaat: 'A Tribute to Vermeer', met een vleugje editing en sound design.
396
907
214
3,664
314,226
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
β€œCoding” was never the source of value, and people shouldn’t get overly attached to it. Problem solving is the core skill. The discipline and precision demanded by traditional programming will remain valuable transferable attributes, but they won’t be a barrier to entry. Many times over the years I have thought about a great programmer I knew that loved assembly language to the point of not wanting to move to C. I have to fight some similar feelings of my own around using existing massive codebases and inefficient languages, but I push through. I had somewhat resigned myself to the fact that I might be missing out on the β€œfinal abstraction”, where you realize that managing people is more powerful than any personal tool. I just don’t like it, and I can live with the limitations that puts on me. I suspect that I will enjoy managing AIs more, even if they wind up being better programmers than I am.
This Post is from an account that no longer exists. Learn more
521
2,510
478
16,515
2,754,163
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
πŸ€– Just published the first 3 episodes of "Hackers Incorporated" β€” me and @r00k's new podcast on everything we're learning about running our businesses. If you're a dev who wants to start their own thing, the value will be off the charts, I promise. hackersincorporated.com/
31
60
17
562
97,990
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
hey gpt4, make me an iPhone app that recommends 5 new movies every day + trailers + where to watch. My ambitions grew as we went along πŸ‘‡
263
1,474
505
11,068
3,994,767
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
oh my god amsterdam looks *incredibly* realistic in the new call of duty β€” almost can’t believe this is a video game
1,617
10,610
3,792
126,791
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
Developers, get to know your terminal! Here are some πŸ”₯ tips for bash and ZSH users that will make you more productive. 🧡
18
170
16
713
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
Iceland: 1, tech industry: 0
320
5,952
1,687
25,291
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
An exciting day! @spolsky called me and let me know. Today's sale of Stack Overflow, most importantly, lets Stack Overflow continue as an independent site -- and also mints 61 new millionaires.
60
154
82
1,589
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
πŸ₯° Gotta love constructor property promotion! #php8 #php If you don’t know the feature yet, here’s an introduction: spatie.be/videos/front-line-…
13
45
12
331
That’s pretty spiffy! Feels like a new paradigm πŸ€”
I keep thinking about this pattern where CSS rules "offer" custom properties for certain config options that can be set inline. It feels very "component'y". Does anybody use this pattern? Are there downsides? stefanjudis.com/notes/how-to…
πŸ„°πŸ„»πŸ„΄πŸ…‡ retweeted
πŸ€”Internet Explorer to be discontinued by Microsoft this August 2021
4
38
7
74