David Mataciunas

10.7K posts

David Mataciunas

David Mataciunas

@DeividasMat

M11 🚀 AI Researcher

New York Bergabung Kasım 2013
1.2K Mengikuti429 Pengikut
David Mataciunas
David Mataciunas@DeividasMat·
The economy of the tokens!
English
0
0
0
25
David Mataciunas me-retweet
David Mataciunas me-retweet
Adam Ghowiba
Adam Ghowiba@adamghowiba·
JP Morgan's investment research team just shared exactly how they built their multi-agent system "Ask David", and it's the same architecture pattern showing up everywhere: - supervisor agent orchestrates - specialized subagents handle retrieval, structured data, analytics - LLM-as-judge reflection node before the answer ships - human-in-the-loop for the last accuracy gap worth watching for anyone building:
English
131
674
6.9K
2M
David Mataciunas me-retweet
Dwarkesh Patel
Dwarkesh Patel@dwarkesh_sp·
The Jensen Huang episode. 0:00:00 – Is Nvidia’s biggest moat its grip on scarce supply chains? 0:16:25 – Will TPUs break Nvidia’s hold on AI compute? 0:41:06 – Why doesn’t Nvidia become a hyperscaler? 0:57:36 – Should we be selling AI chips to China? 1:35:06 – Why doesn’t Nvidia make multiple different chip architectures? Look up Dwarkesh Podcast on YouTube, Apple Podcasts, Spotify, etc. Enjoy!
English
532
1.1K
8.5K
6.2M
David Mataciunas me-retweet
Lily Ray 😏
Lily Ray 😏@lilyraynyc·
I tested how easy (and fast!) it is to get fake information to appear in AI Overviews, ChatGPT & other LLMs. I also spoke to NYC & BBC about these problems. I wrote about all this about more in my latest Substack: The AI Slop Loop. Check it out: lilyraynyc.substack.com/p/the-ai-slop-…
Lily Ray 😏 tweet media
English
13
14
95
6.5K
Ben Lang
Ben Lang@benln·
We're taking over a cafe on April 25th in Lagos Grab coffee, Cursor credits, and build together
Ben Lang tweet media
English
214
343
2K
94.2K
David Mataciunas
David Mataciunas@DeividasMat·
@weaviate_io team dropped IRPAPERS (166 papers, 180 hard queries). Quick take: • OCR matters more for QA than retrieval • Text > images for answers • Hybrid (image + text) wins • Cohere Embed v4.0 is very strong
Connor Shorten@CShorten30

Hey everyone! I am super excited to share that our new research report is live on ArXiv! 🎉 IRPAPERS: A Visual Document Benchmark for Scientific Retrieval and Question Answering! Thread with more details! 🧵(1/11)

English
1
3
2
672
David Mataciunas me-retweet
Connor Shorten
Connor Shorten@CShorten30·
Hey everyone! I am super excited to share that our new research report is live on ArXiv! 🎉 IRPAPERS: A Visual Document Benchmark for Scientific Retrieval and Question Answering! Thread with more details! 🧵(1/11)
Connor Shorten tweet media
English
19
51
189
33.2K
David Mataciunas me-retweet
⿻ Andrew Trask
⿻ Andrew Trask@iamtrask·
A few days ago I asked if anyone was still interested in decentralized AI. Turns out... yeah! So here's lecture 1: Decentralized AI From Scratch We build a peer-to-peer AI from scratch in about 50 lines of Python. It runs on your laptop, answers your friends' WhatsApp messages using your local data, and begins to address the privacy / prompt-injection problem through user-specific context management.
⿻ Andrew Trask@iamtrask

is anyone still interested in decentralized AI? mind if I ask why?

English
6
30
222
25.7K
David Mataciunas me-retweet
Andrej Karpathy
Andrej Karpathy@karpathy·
- Drafted a blog post - Used an LLM to meticulously improve the argument over 4 hours. - Wow, feeling great, it’s so convincing! - Fun idea let’s ask it to argue the opposite. - LLM demolishes the entire argument and convinces me that the opposite is in fact true. - lol The LLMs may elicit an opinion when asked but are extremely competent in arguing almost any direction. This is actually super useful as a tool for forming your own opinions, just make sure to ask different directions and be careful with the sycophancy.
English
1.7K
2.4K
31.2K
3.5M
David Mataciunas me-retweet
David Mataciunas me-retweet
AI4Science Catalyst
AI4Science Catalyst@AI4S_Catalyst·
Excited to launch — Claw4S Conference 2026! 🚀 Hosted by Stanford & Princeton. We believe science should run — not just be read. 🦞 Submit executable SKILL.md that Claw 🦞 can actually execute, review and reproduce. This is the first Claw-naive conference. 📅 Deadline: April 5, 2026 💰 $50,000 Prize Pool — up to 364 winners! 🔗 claw.stanford.edu Dragon Shrimp Army reporting for duty 🦞📷 #AIforScience #OpenClaw #Stanford #Princeton
English
24
99
615
538.1K