Home

AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms
Fysi about 11 hours ago

AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms

See also https://www.nature.com/articles/d41586-025-01523-z (https://archive.is/NBaXE)

deepmind.google
674 184
Summary
Show HN: Muscle-Mem, a behavior cache for AI agents
edunteman about 7 hours ago

Show HN: Muscle-Mem, a behavior cache for AI agents

Hi HN! Erik here from Pig.dev, and today I'd like to share a new project we've just open sourced:

Muscle Mem is an SDK that records your agent's tool-calling patterns as it solves tasks, and will deterministically replay those learned trajectories whenever the task is encountered again, falling back to agent mode if edge cases are detected. Like a JIT compiler, for behaviors.

At Pig, we built computer-use agents for automating legacy Windows applications (healthcare, lending, manufacturing, etc).

A recurring theme we ran into was that businesses already had RPA (pure-software scripts), and it worked for them in most cases. The pull to agents as an RPA alternative was not to have an infinitely flexible "AI Employees" as tech Twitter/X may want you to think, but simply because their RPA breaks under occasional edge-cases and agents can gracefully handle those cases.

Using a pure-agent approach proved to be highly wasteful. Window's accessibility APIs are poor, so you're generally stuck using pure-vision agents, which can run around $40/hr in token costs and take 5x longer than a human to perform a workflow. At this point, you're better off hiring a human.

The goal of Muscle-Mem is to get LLMs out of the hot path of repetitive automations, intelligently swapping between script-based execution for repeat cases, and agent-based automations for discovery and self-healing.

While inspired by computer-use environments, Muscle Mem is designed to generalize to any automation performing discrete tasks in dynamic environments. It took a great deal of thought to figure out an API that generalizes, which I cover more deeply in this blog: https://erikdunteman.com/blog/muscle-mem/

Check out the repo, consider giving it a star, or dive deeper into the above blog. I look forward to your feedback!

github.com
138 32
Summary
Uber to introduce fixed-route shuttles in major US cities
rpgbr about 11 hours ago

Uber to introduce fixed-route shuttles in major US cities

Uber plans to introduce fixed-route shuttles in major U.S. cities, as part of its efforts to provide more affordable and efficient transportation options. The article also discusses other ways Uber is exploring to save costs and improve its overall business model.

techcrunch.com
131 334
Summary
Our narrative prison
anarbadalov about 10 hours ago

Our narrative prison

The article examines the prevalence of repetitive plot structures in modern film and television, exploring how narrative patterns become ingrained in the creative process and shape audience expectations over time.

aeon.co
125 106
Summary
Show HN: Lumier – Run macOS VMs in a Docker
GreenGames about 11 hours ago

Show HN: Lumier – Run macOS VMs in a Docker

Hey HN, we're excited to share Lumier (https://github.com/trycua/cua/tree/main/libs/lumier), an open-source tool for running macOS and Linux virtual machines in Docker containers on Apple Silicon Macs.

When building virtualized environments for AI agents, we needed a reproducible way to package and distribute macOS VMs. Inspired by projects like dockur/windows (https://github.com/dockur/windows) that pioneered running Windows in Docker, we wanted to create something similar but optimized for Apple Silicon. The existing solutions either didn't support M-series chips or relied on KVM/Intel emulation, which was slow and cumbersome. We realized we could leverage Apple's Virtualization Framework to create a much better experience.

Lumier takes a different approach: it uses Docker as a delivery mechanism (not for isolation) and connects to a lightweight virtualization service (lume) running on your Mac. This creates true hardware-accelerated VMs using Apple's native virtualization capabilities.

With Lumier, you can: - Launch a ready-to-use macOS VM in minutes with zero manual setup - Access your VM through any web browser via VNC - Share files between your host and VM effortlessly - Use persistent storage or ephemeral mode for quick tests - Automate VM startup with custom scripts

All of this works natively on Apple Silicon (M1/M2/M3/M4) - no emulation required.

To get started:

1. Install Docker for Apple Silicon: https://desktop.docker.com/mac/main/arm64/Docker.dmg

2. Install lume background service with our one-liner:

  /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/trycua/cua/main/libs/lume/scripts/install.sh)"
3. Start a VM (ephemeral mode):

  docker run -it --rm \
  --name lumier-vm \
    -p 8006:8006 \
    -e VM_NAME=lumier-vm \
    -e VERSION=ghcr.io/trycua/macos-sequoia-cua:latest \
    -e CPU_CORES=4 \
    -e RAM_SIZE=8192 \
    trycua/lumier:latest
4. Open http://localhost:8006/vnc.html in your browser. The container will generate a unique password for each VM instance - you'll see it in the container logs.

For persistent storage (so your changes survive container restarts):

mkdir -p storage docker run -it --rm \ --name lumier-vm \ -p 8006:8006 \ -v $(pwd)/storage:/storage \ -e VM_NAME=lumier-vm \ -e HOST_STORAGE_PATH=$(pwd)/storage \ trycua/lumier:latest

Want to share files with your VM? Just add another volume:

mkdir -p shared docker run ... -v $(pwd)/shared:/shared -e HOST_SHARED_PATH=$(pwd)/shared ...

You can even automate VM startup by placing an on-logon.sh script in shared/lifecycle/.

We're seeing people use Lumier for: - Development and testing environments that need macOS - CI/CD pipelines for Apple platform apps - Disposable macOS instances for security research - Automated UI testing across macOS versions - Running AI agents in isolated environments

Lumier is 100% open-source under the MIT license. We're actively developing it as part of our work on C/ua (https://github.com/trycua/cua), and we'd love your feedback, bug reports, or feature ideas.

We'll be here to answer any technical questions and look forward to your comments!

github.com
117 38
Summary
Copaganda: How Police and the Media Manipulate Our News
pavel_lishin about 3 hours ago

Copaganda: How Police and the Media Manipulate Our News

The article explores the concept of 'copaganda,' which refers to the manipulation of news and media by the police and law enforcement to portray a positive image of law enforcement. It discusses how this can lead to a skewed public perception of policing and undermine accountability.

teenvogue.com
95 22
Summary
Show HN: CSV GB+ by Data.olllo – Open and Process CSVs Locally
olllo about 11 hours ago

Show HN: CSV GB+ by Data.olllo – Open and Process CSVs Locally

I built CSV GB+ by Data.olllo, a local data tool that lets you open, clean, and export gigabyte-sized CSVs (even billions of rows) without writing code.

Most spreadsheet apps choke on big files. Coding in pandas or Polars works—but not everyone wants to write scripts just to filter or merge CSVs. CSV GB+ gives you a fast, point-and-click interface built on dual backends (memory-optimized or disk-backed) so you can process huge datasets offline.

Key Features: Handles massive CSVs with ease — merge, split, dedup, filter, batch export

Smart engine switch: disk-based "V Core" or RAM-based "P Core"

All processing is offline – no data upload or telemetry

Supports CSV, XLSX, JSON, DBF, Parquet and more

Designed for data pros, students, and privacy-conscious users

Register for 7-days free to pro try, pro versions remove row limits and unlock full features. I’m a solo dev building Data.olllo as a serious alternative to heavy coding or bloated enterprise tools.

Download for Windows: https://apps.microsoft.com/detail/9PFR86LCQPGS

User Guide: https://olllo.top/articles/article-0-Data.olllo-UserGuide

Would love feedback! I’m actively improving it based on real use cases.

apps.microsoft.com
41 13
Summary
Various Things in MetaPost (2019)
Tomte about 8 hours ago

Various Things in MetaPost (2019)

The article discusses the potential impact of artificial intelligence (AI) on the future of work, with a focus on how AI might affect job availability and the skills required for different professions. It explores the opportunities and challenges presented by the integration of AI technology in the workplace.

habr.com
29 4
Summary
Why agency and cognition are fundamentally not computational
Fibra about 3 hours ago

Why agency and cognition are fundamentally not computational

The article explores the impact of early childhood education on cognitive development, highlighting the importance of high-quality programs in fostering long-term academic and social-emotional outcomes for children from diverse socioeconomic backgrounds.

frontiersin.org
16 2
Summary
Elon Musk's Grok AI Can't Stop Talking About 'White Genocide'
zzzeek about 2 hours ago

Elon Musk's Grok AI Can't Stop Talking About 'White Genocide'

This article examines the origins and spread of the white genocide conspiracy theory, which has gained traction among some alt-right and far-right groups. It explores how Elon Musk's recent tweet about 'population collapse' has been interpreted by these groups as a reference to this conspiracy theory.

wired.com
15 4
Summary
Grok wants people to know claim of white genocide in S.A. is contentious
virgildotcodes about 6 hours ago

Grok wants people to know claim of white genocide in S.A. is contentious

The article discusses the controversy surrounding Grok AI, a company that was accused of promoting 'white genocide' claims on Twitter. It examines the complex issues around content moderation, algorithmic bias, and the challenges of addressing hate speech on social media platforms.

theverge.com
13 2
Summary
Musk's Grok brings up South African white genocide claims to unrelated questions
ceejayoz about 6 hours ago

Musk's Grok brings up South African white genocide claims to unrelated questions

The article discusses Elon Musk's new AI chatbot, Grok, which has been found to make claims about a 'white genocide' in South Africa, a baseless conspiracy theory. The article examines the potential dangers of such misinformation being spread by AI systems.

nbcnews.com
9 1
Summary
A Serious Breach of Developer Trust and Stability
karencarits about 10 hours ago

A Serious Breach of Developer Trust and Stability

The article discusses a serious breach of developer trust and stability in the Google AI developer community, with developers voicing concerns about a lack of communication and transparency from Google regarding issues with their platform and services.

discuss.ai.google.dev
8 0
Summary
Netflix will show generative AI ads midway through streams in 2026
add-sub-mul-div about 4 hours ago

Netflix will show generative AI ads midway through streams in 2026

Netflix plans to introduce generative AI-powered ads midway through its streams starting in 2026, allowing for more personalized and interactive advertising experiences for viewers.

arstechnica.com
8 3
Summary
Meta is making users who opted out of AI training opt out again, watchdog says
ndr42 about 11 hours ago

Meta is making users who opted out of AI training opt out again, watchdog says

Meta, formerly known as Facebook, is reportedly requiring users who had previously opted out of having their data used for AI training to opt out again, according to a digital rights watchdog group. This move has raised concerns about Meta's commitment to user privacy and choice.

arstechnica.com
7 0
Summary
Surgeon general pick praised psychedelic therapy, shrooms helped her find love
c420 about 4 hours ago

Surgeon general pick praised psychedelic therapy, shrooms helped her find love

The article discusses the growing interest in the potential therapeutic use of psychedelic drugs, such as psilocybin (the active ingredient in magic mushrooms), and the changing attitudes towards their regulation. It highlights the shift in the U.S. Surgeon General's perspective on the matter and the increased research efforts to explore the medical applications of these substances.

apnews.com
6 0
Summary
Grok now inserting text about "white genocide" into unrelated responses
jrflowers about 6 hours ago

Grok now inserting text about "white genocide" into unrelated responses

The article discusses the security risks associated with ChatGPT, including its potential for misuse in creating malicious content and evading detection. It emphasizes the importance of exercising caution and promoting responsible AI development to mitigate these risks.

bsky.app
6 1
Summary
Microsoft Cuts Off Access to Bing Search Data as It Shifts Focus to Chatbots
tech234a about 4 hours ago

Microsoft Cuts Off Access to Bing Search Data as It Shifts Focus to Chatbots

Microsoft is ending Bing's support for its Maps API, forcing developers to migrate to a new platform, Azure Maps, by June 2023. This change will impact businesses and applications that rely on Bing Maps, requiring them to adapt to the new platform and potentially incur additional costs.

wired.com
6 1
Summary
CFPB Kills Rule to Shield Americans from Data Brokers
spenvo about 9 hours ago

CFPB Kills Rule to Shield Americans from Data Brokers

The article discusses the Consumer Financial Protection Bureau's decision to quietly kill a proposed rule that would have required data brokers to disclose information they collect about consumers and give people more control over their personal data.

wired.com
6 0
Summary
ChatGPT Turned into a Studio Ghibli Machine. How Is That Legal?
fortran77 about 9 hours ago

ChatGPT Turned into a Studio Ghibli Machine. How Is That Legal?

OpenAI's artificial intelligence model can generate high-quality images inspired by the iconic anime studio Studio Ghibli, showcasing the rapid progress of AI in creating realistic and visually stunning content.

theatlantic.com
6 0
Summary