2581
New
The latest blog posts to take off on Hacker News. 6,331 posts.
2582
Beyond Diffusion: Inductive Moment Matching
Inductive Moment Matching surpasses diffusion models in speed and sample quality.
2583
Happy 10k Day
Today, we sold the 10,000th comma 3X, our first product breaking 5 digits in sales.
2584
2585
People are just as bad as my LLMs
Last year I created a fun little experiment where I asked a bunch of LLMs to rank 97 hackernews users using their comment history based on whether they would be good candidates for the role of “software engineer at google”. (yes yes, seems silly I know, you can read part 1 and part 2 but […]
2586
Generative AI hype peaking?
I have been a skeptic of some of the more breathless claims made for Generative AI since at least October 2023, when I participated in a panel organized by S...
2587
With AI you need to think bigger
I have noticed something in the past 12 months, something a little profound. I have noticed that I am no longer scared that a project will be too big or too complex for me, or that a project will use a technology or programming language I don’t know.
2588
MCP vs. API Explained
Model Context Protocol (MCP) is an open standard that connects AI models to tools and data sources efficiently. This guide breaks down MCP’s architecture, benefits, and how it differs from traditional APIs
2589
The Einstein AI Model
I shared a controversial take the other day at an event and I decided to write it down in a longer format: I’m afraid AI won't give us a compressed 21st century.
2590
Long Read: Lessons from Building Semantic Search for GitHub and Why I Failed
Made in Notion, the connected workspace where AI agents and teams organize docs, projects and knowledge together.
2591
AMD YOLO
AMD is sending us the two MI300X boxes we asked for. They are in the mail.
2592
Why I find diffusion models interesting?
HN Discussion
2593
Using GRPO to Beat o1, o3-mini and R1 at “Temporal Clue”
The OpenPipe platform has migrated to Weights & Biases and CoreWeave. Browse migration details and the OpenPipe technical archive.
2594
Mistral OCR
The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.
2595
Automatically tagging politician when they use their phone on the livestreams
How does it work? Every meeting of the flemish government in Belgium is live streamed on a youtube channel. When a livestream starts the software is searching for phones and tries to identify a distracted politician. This is done with the help of AI and face recognition. The video of the distracted politician are then…
2596
QwQ-32B: Embracing the Power of Reinforcement Learning
QWEN CHAT Hugging Face ModelScope DEMO DISCORD Scaling Reinforcement Learning (RL) has the potential to enhance model performance beyond conventional pretraining and post-training methods. Recent studies have demonstrated that RL can significantly improve the reasoning capabilities of models. For instance, DeepSeek R1…
2597
Skynet won and destroyed humanity
For a very long time, it appeared obvious to outside observers of the time flow that Skynet would never win the war, and never annihilate humanity.
2598
Richard Sutton and Andrew Barto Win 2024 Turing Award
Andrew Barto and Richard Sutton as the recipients of the 2024 ACM A.M. Turing Award for developing the conceptual and algorithmic foundations of reinforcement learning. In a series of papers beginning in the 1980s, Barto and Sutton introduced the main ideas, constructed the mathematical foundations, and developed…
2599
Writing an LLM from scratch, part 8 – trainable self-attention
Moving on from a toy self-attention mechanism, it's time to find out how to build a real trainable one. Following Sebastian Raschka's book 'Build a Large Language Model (from Scratch)'. Part 8/??
2600
ARC-AGI without pretraining
We solve 20% of ARC-AGI-1 with no pretraining by minimizing the description length during inference time, with the only data involved being the target puzzle itself.
2601
2602
DeepSeek's smallpond: Bringing Distributed Computing to DuckDB
DeepSeek is pushing DuckDB beyond its single-node roots with smallpond, a new, simple approach to distributed compute. But does it solve the scalability challenge—or introduce new trade-offs?
2603
The Differences Between Deep Research, Deep Research, and Deep Research
Dive into the world of AI engineering with this exploration of Deep Research in report generation. Learn how LLM-as-a-judge systems, machine learning, and ev...
2604
Hallucinations in code are the least dangerous form of LLM mistakes
A surprisingly common complaint I see from developers who have tried using LLMs for code is that they encountered a hallucination—usually the LLM inventing a method or even a full …
2605
GPT-4.5: "Not a frontier model"?
OpenAI's latest model raises more questions than answers, but no, the AI bubble isn't popping quite yet.
2606
Crossing the uncanny valley of conversational voice
Voice is our most intimate medium as humans. At Sesame, we are creating conversational partners that achieve voice presence—the quality that makes spoken interactions feel real, understood, and valued.
2607
Making o1, o3, and Sonnet 3.7 hallucinate for everyone
My (ops-but-sometimes-writes-scripts-to-help-out) coworker just tapped on my shoulder and asked me to look at his code that wasn't working. It was a bit something like this:
2608
Yes, Claude Code can decompile itself. Here's the source code
These LLMs are shockingly good at deobfuscation, transpilation and structure to structure conversions. I discovered this back around Christmas where I asked an LLM to make me an Haskell audio library by transpiling a rust implementation. An “oh fuck” moment in timeOver the Christmas break I’ve been critically looking
2609
AI is killing some companies, yet others are thriving – let's look at the data
Traffic trends from WebMD, Quora, Stack Overflow, Chegg, G2, CNET, Reddit, Wikipedia, and Substack!
2610
Hot take: GPT 4.5 is a nothing burger
Pure scaling in shambles