TLDR 2025-07-23
iPhone Fold leaks π±, OpenAI's $30B/yr deal π°, high performance vibe coding π¨βπ»Β
π Delve Raises $32M Series A to change compliance forever (Sponsor)
We're thrilled to announce our
$32M Series A at a $300M valuation, led by Insight Partners!
Delve is shaping the future of compliance with an AI-native approach that cuts busywork and saves teams hundreds of hours. Startups like Lovable, Bland, and Browser trust our AI to get compliantβfast.
To celebrate, we're giving back with 3 limited-time offers:
- $15,000 referral bonus if you refer a founding engineer we hire
- $2,000 off compliance setup for new customers
- A custom Delve doormat for anyone who reposts + comments on our LinkedIn post (while supplies last!)
Thank you for your supportβthis is just the beginning.
Get started with DelveBook your demo now
OpenAI agreed to pay Oracle $30B a year for data center services (2 minute read)
Oracle disclosed in an SEC filing last month that it had signed a cloud deal that would generate $30 billion a year in revenue. The company that signed that deal has been revealed to be OpenAI. Oracle collectively sold $24.5 billion worth of cloud services in its fiscal 2025 to all customers combined. OpenAI and Oracle still have to build the data center, which will be a costly endeavor.
Foldable iPhone's Display Sizes Leaked (1 minute read)
Apple will likely launch its foldable iPhone in the second half of 2026. The device will come with a 7.8-inch inner display and a 5.5-inch outer display. The Samsung Galaxy Z Fold 7 has an 8-inch inner screen and a 6.5-inch outer screen. Apple's foldable iPhone will feature a 'crease-free' inner display and have two rear cameras, one front camera, and a Touch ID power button instead of Face ID.
π
Science & Futuristic Technology
Genetics Reimagined (4 minute read)
Scientists recently confirmed the healthy births of children through mitochondrial donation. The innovation opens new doors for reproductive choice. It offers at-risk families a viable path to parenthood without the lingering fear of passing on devastating inherited conditions. However, it also opens up the possibility of designer genetics that will be passed down to future generations. Those currently living with conditions will still face daily challenges, but for the next generation, the landscape has changed.
DeepMind's Quest for Self-Improving Table Tennis Agents (5 minute read)
Google DeepMind recently used table tennis as a testbed for its technology, as the sport requires robots to master a confluence of difficult skills - it requires perception and demands exceptionally precise control. This makes table tennis an ideal domain for developing and evaluating robust learning algorithms that can handle real-time interaction, complex physics, high-level reasoning, and the need for adaptive strategies. This article looks at the study and the insights gained from it. One important lesson learned from the project is that VLMs can be leveraged for explainable robot policy search.
π»
Programming, Design & Data Science
any-agent (GitHub Repo)
any-agent is a Python library that provides a single interface for evaluating different agent frameworks. It can be used to create and evaluate agents, implement and use custom callbacks, integrate Model Context Protocol tools, deploy agents with Agent-to-Agent communication, and more. Tutorials on how to get started are available.
Vibecoding a high performance system (24 minute read)
This developer recently used agentic coding to build a system to crawl a billion web pages in around 24 hours. Less than 4% of the code was written by hand. The post discusses how AI helped code the tool and where it fell short. AI was a huge boost during the development process. Some of the learnings from this project could also apply to building other high-performance systems.
Reading QR codes without a computer! (9 minute read)
This article provides an interactive explanation of how QR codes work. It covers the anatomy of a QR code, mask patterns, encoding, reading, and decoding. Readers can enter their own custom data to generate a QR code to follow along with the article. The article gives readers all the information they need to decode simple QR codes by hand.
How Apple's Internal Veto Against Open-Source AI Sparked a Talent Exodus to Meta (4 minute read)
Apple had a plan to open-source its core AI models, but its leadership decided against it due to fears of exposing performance compromises made for on-device processing, a pivotal decision that has ignited a crisis of confidence within the company's AI division. The move has directly triggered an exodus of top talent to rivals like Meta. Many engineers are frustrated with the internal conflict between Apple's famous secrecy and the open culture of AI research. There are now serious doubts over the future of Apple Intelligence and the company's ability to compete in the AI arms race.
Get the most interesting stories in startups, tech, and programming delivered in a free daily email.
Join 1,600,000 readers for
one daily email