Hosting a website on a disposable vape
bogdanthegeek.github.io
Massive Attack Turns Concert into Facial Recognition Surveillance Experiment
gadgetreview.com
William Gibson Reads Neuromancer
bearcave.com
Addendum to GPT-5 system card: GPT-5-Codex
openai.com
React is winning by default and slowing innovation
lorenstew.art
Wanted to spy on my dog, ended up spying on TP-Link
kennedn.com
PayPal to support Ethereum and Bitcoin
newsroom.paypal-corp.com
Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
How big a solar battery do I need to store all my home's electricity?
shkspr.mobi
When Your Father Is a Magician, What Do You Believe?
thereader.mitpress.mit.edu
How People Use ChatGPT [pdf]
cdn.openai.com
CubeSats are fascinating learning tools for space
jeffgeerling.com
Show HN: Pooshit – sync local code to remote Docker containers
Scryer Prolog Meetup 2025
hsd-pbsa.de
How to self-host a web font from Google Fonts
blog.velocifyer.com
I wish my web server were in the corner of my room (2022)
interconnected.org
Boring work needs tension
iaziz786.com
GPT‑5-Codex and upgrades to Codex
simonwillison.net
Removing newlines in FASTA file increases ZSTD compression ratio by 10x
log.bede.im
RustGPT: A pure-Rust transformer LLM built from scratch
github.com
Hi HN,
We just launched Blocks - an AI platform where anyone can build custom work apps and AI agents in minutes, no coding required.
We’re seeing a shift in how software is built and used. With AI, software isn’t just about tracking processes anymore - it can actively help people get work done.
The problem is that most teams still face 2 options: either buy off-the-shelf tools that never quite fit, or invest a lot of time and money into custom development. Neither works well.
Blocks is the third option. It’s an AI-native platform where teams can create apps and AI agents for their workflows just by describing what they need in plain language. The idea is simple: the person who feels the problem should be able to build the solution.
Each app in Blocks has three layers that work together:
UI: customizable interfaces for each role.
AI agents: automate tasks, integrate with systems, and take action.
Data: a built-in database
Having all three in one place means you don’t just get a custom tool - you get a system that automates, adapts, and improves over time. Teams can collaborate internally and externally, set roles and permissions, and connect to the systems they already use.
How it works: you describe what you need in plain language, and Ella (our AI builder) creates the app - UI, agents, and database. From there you can edit, extend, and integrate with tools like Google Sheets, Slack, HubSpot, monday.com and many more. Custom integrations are rolling out soon.
We’d love for you to check it out → https://blocks.diy
It’s still early days, so your feedback would be hugely valuable. We’d really appreciate feedback on:
How clear the build process feels to non-technical users.
Which integrations or agent use cases would be most useful.
Any rough edges you notice while testing.
Thanks for reading! we’re excited to hear your thoughts and happy to answer any questions.