Search the archive

HOW TO MAKE SUPERBABIES

Genetic technology could create humans that live longer, have the raw intellectual power to do Nobel Prize-worthy scientific research, and very rarely suffer from health disorders. Scientists don't do this though due to ethical issues. As a result, there is a lot of low-hanging fruit in this field of research. This post looks into how to do large-scale, heritable editing of the human genome to improve everything from diabetes risk to intelligence. It summarizes the state of knowledge in the field and lays out a technical roadmap that examines how remaining barriers may be overcome.

EARTH'S 1ST ASTEROID MINING PROSPECTOR HEADS TO THE LAUNCHPAD

AstroForge is planning to heave a microwave oven-sized spacecraft toward an asteroid later this week as part of the company's plans to create a future where precious metals are mined around the solar system. Its Odin spacecraft will launch on a SpaceX Falcon 9 rocket as soon as Wednesday. Odin's target is an object called 2022 OB5, which is about the size of a football field and is likely to contain a high amount of metal. Odin will arrive in late 2025 after a journey of about 300 days.

it here

HOW TO BUILD AN LLM CHAT APP: THE NEW LITMUS TEST FOR JUNIOR DEVS

Building an LLM (Large Language Model) chat app is way more complicated than just making an API call. There are many considerations required - building an LLM chat app is the gauntlet that separates the juniors from the devs who actually know what they are doing. This article discusses how to start building a scalable LLM chat app. Learning how to build the backend architecture allows developers to build better apps that can handle more users and dodge the shame of a launch-day crash.

CSS ZERO

CSS Zero is an opinionated CSS starter kit for Ruby on Rails applications. It is akin to a 'no build' Tailwind CSS. CSS Zero includes custom templates for scaffolds and authentication.

WHY I THINK AI TAKE-OFF IS RELATIVELY SLOW

A large part of the economy just won't adopt AI or use it effectively all that quickly. As AI becomes more productive, human bottlenecks will become more important. Not all bottlenecks will be regulatory: there will be slow adopters, people in the workplace who hate AI, energy constraints, and other factors. It is the worst performer who sets the overall level of productivity - soon, the worst performer will be humans. While AI could boost economic rates, that reality will still be a few decades out, if it happens.

THE PROFITABLE STARTUP

The thinking that startups should prioritize growth over everything else, even profitability, was always flawed - profitability isn't unambitious. It means not relying on investors for survival and being able to focus on an unaltered vision and mission. It's now easy to be profitable while also growing fast. The factor holding founders back is rarely team size - it's the clarity of their focus, skill, and ability to execute. Measure what matters, understand your risk profile, hire intentionally and slowly, and raise on your own terms.

GOOGLE'S NEW AI VIDEO MODEL VEO 2 WILL COST 50 CENTS PER SECOND

Google's new video-generating AI model can create clips that are two minutes or more.

DID XAI LIE ABOUT GROK 3'S BENCHMARKS?

The methods that AI labs use to present AI benchmark results can be questionable.

UBER CEO SAYS MUSK WANTS TO GO ALONE ON TESLA ROBOTAXI ROLLOUT

Uber, Waymo, and Tesla will be competing against each other when Tesla launches its robotaxis in Austin.

THE IPV6 TRANSITION

The transition to IPv6 has progressed very slowly because there's no real lasting benefit in trying to leap across to another 1980s address-based architecture with only a few differences.

APPLE'S C1 MODEM IS THE FIRST STEP TOWARDS 'A PLATFORM FOR GENERATIONS,' EXECUTIVES SAY

Apple's C1 modem, which lacks mmWave support but features custom GPS and satellite connectivity, was designed to cater specifically to the needs of the iPhone 16e and future devices.

ELON MUSK'S SPACEX IS GETTING READY TO TAKE THE INTERNATIONAL SPACE STATION OUT OF ORBIT

NASA has awarded SpaceX a contract to deorbit the International Space Station no earlier than 2030.

February 27.

OPENAI'S GPT-4.5 MAY ARRIVE NEXT WEEK, BUT GPT-5 IS JUST AROUND THE CORNER

OpenAI may be preparing to release GPT-4.5 as early as next week - anonymous sources say that the next model could hit the scene sometime this month. GPT-5 may be released as early as May. Microsoft has a good reason to want GPT-5 in May - its annual developer conference is on May 22. While GPT-5 may be more capable and efficient, it doesn't necessarily mean there will be any new use cases for AI.

AMAZON MGM STUDIOS GAINS CREATIVE CONTROL OF JAMES BOND FRANCHISE FROM BROCCOLI FAMILY

The Broccoli family, which has long controlled the 007 franchise, will hand over creative control of the series to Amazon MGM Studios in a new joint venture with the studio. Amazon will now control who will play Bond, who will write the next script, and when the film goes into production. The financial terms of the joint venture have not been disclosed. Amazon and the Broccoli family have been at odds since the tech company acquired the rights to release Bond movies about three years ago.

BIGGEST-EVER AI BIOLOGY MODEL WRITES DNA ON DEMAND

Scientists have released an artificial intelligence model for biology trained on 128,000 genomes spanning the tree of life. It can make sense of existing DNA and also write whole chromosomes and small genomes from scratch. The model is available to scientists through a web interface - it can also be downloaded freely. The model could be used to help identify hard-to-interpret changes in people's genomes, decipher complex genomes, predict the effects of mutations, and more.

FIGURE'S HUMANOID ROBOT TAKES VOICE ORDERS TO HELP AROUND THE HOUSE

Figure has revealed a Vision-Language-Action (VLA) model for humanoid robots. VLAs leverage vision and language commands to process information, enabling robots to be trained through a combination of video and large language models. Figure's Helix model enables robots to follow natural language commands and pick up novel items with varying shapes, sizes, colors, and material properties never encountered before in training. Work on Helix is still at a very early stage, so demonstrations should be taken with a grain of salt.

DEEPEVAL

DeepEval is an open source framework for evaluating and testing large language model systems. It incorporates the latest research to evaluate model outputs using locally run models. DeepEval supports applications implemented via RAG, fine-tuning, LangChain, LlamaIndex, and more. It helps developers easily determine the optimal hyperparameters to improve their RAG pipeline, improve prompt drifting, or transition from OpenAI to hosting their own Llama 3.

KUBEVPN

KubeVPN provides a Cloud-Native Dev Environment that seamlessly connects to Kubernetes cluster networks using service names or Pod IP/Service IP. It enables developers to run Kubernetes pods within local Docker containers, ensuring an identical environment, volume, and network setup. KubeVPN empowers developers to create applications entirely on their local PCs.

SOFTWARE ENGINEERING JOB OPENINGS HIT FIVE-YEAR LOW?

There has been a 35% decrease in the number of software engineering job vacancies in the US since January 2020. While the story is similar in Canada, things are different in the UK, France, Germany, and Australia. Australia is the only country where the number of jobs listed is not lower than in 2020. Software development jobs have been the biggest boom and bust in vacancies compared to all industries. This post discusses the reasons for the decrease and predictions for what growth in the tech industry is likely to be like this year.

A.I. IS CHANGING HOW SILICON VALLEY BUILDS START-UPS

Tech startups used to raise huge sums to hire armies of workers and grow fast, but AI tools are enabling workers to be more efficient, making tiny teams more successful. Startups are using AI tools to increase employees' productivity in everything from customer service and marketing to coding and customer research. They are able to gain more revenue with fewer employees and less cash, something that wouldn't have been possible without the technology. The potential for AI to let startups do more with less has led to wild speculation about the future.

AWS S3 SDK BREAKS ITS COMPATIBLE SERVICES

The new default data protections for new objects in Amazon S3 break compatibility with nearly all S3-compatible services as they update the default settings in all SDKs.

GENSX

GenSX is a TypeScript framework for billing agents and workflows with simple and reusable React-like components.

META STARTS ACCEPTING SIGN-UPS FOR COMMUNITY NOTES ON FACEBOOK, INSTAGRAM, AND THREADS

Community Notes will be a way for Meta users to decide when posts are misleading and add more context to posts.

GOOGLE'S AI CO-SCIENTIST IS 'TEST-TIME SCALING' ON STEROIDS. WHAT THAT MEANS FOR RESEARCH

Google's Co-scientist is a hypothesis-formulation machine that uses multiple agents.

THE SCALING LAWS GOVERN AI PROGRESS

Grok 3 is a massive leap forward compared to Grok 2 - it performs at a level compared to, and in some cases exceeding, models from OpenAI, Google, DeepMind, and Anthropic.

HELLO, DEEPSEEK OPEN INFRA!

DeepSeek is open sourcing five repositories that contain the building blocks of its online service to share its progress with full transparency.

Secure your spot now and earn up to 100% bonus shares.

APPLE LAUNCHES NEW IPHONE 16E: HERE'S EVERYTHING YOU NEED TO KNOW

Apple's iPhone 16e features an all-new design, an A18 chip, a 48-megapixel camera, full Apple Intelligence support, and the first-ever Apple-designed 5G cellular modem, the C1 chip. It starts at $599 for 128GB and will be available for pre-order tomorrow. The first orders will arrive on February 28. Apple has discontinued the iPhone SE along with the iPhone 14 Plus. The iPhone 14 will remain available from Apple's channel partners but it has been removed from Apple's website.

MICROSOFT DEPLOYS NEW STATE OF MATTER IN ITS FIRST QUANTUM COMPUTING CHIP

Microsoft's first quantum computing chip, Majorana 1, is the result of nearly two decades of research in the field. Microsoft claims that building the chip required the creation of an entirely new state of matter called the topological state. Microsoft manufactured all the parts for the chip in the US, which was possible because of the small scale of the project. Quantum computing may end up being used to boost Microsoft's other businesses rather than just being a stand-alone category.

MIGRAINE IS MORE THAN A HEADACHE — A RADICAL RETHINK OFFERS HOPE TO ONE BILLION PEOPLE

Progress in understanding migraines has started to advance beyond focusing on just headaches. Researchers are now developing measures for the cognitive symptoms of migraines. Measuring cognitive symptoms is particularly important to show whether treatments are effective or not. Previous treatment studies focused on headache severity, light sensitivity, and nausea, but little else - cognitive symptoms are commonly felt in between migraine attacks as well. It may require a serious rethink about the causes of migraines in order to make it possible to treat all the complexities of the condition.

NEW BOOK-SORTING ALGORITHM ALMOST REACHES PERFECTION

Researchers have created a new algorithm that addresses the library sorting problem: a challenge to devise a strategy for organizing books in some kind of sorted order that minimizes how long it takes to place a new book on the shelf. The problem applies to the arrangement of files on hard drives and in databases. An inefficient system means significant wait times and major computational expense. The new algorithm is very close to the theoretical ideal.

new VR headset. Register for free

MASTRA

Mastra is an opinionated TypeScript framework that helps developers quickly build AI applications and features. It provides a set of primitives - workflows, agents, RAG, integrations, and evals. Mastra can be run locally or deployed to a serverless cloud.

POSTGRES LANGUAGE SERVER

This repository contains a collection of language tools and a Language Server Protocol implementation for Postgres that focuses on developer experience and reliable SQL tooling. It provides a toolchain for Postgres development built on a Server-Client architecture with a transport-agnostic design. All features can be accessed through the Language Server Protocol and also through other interfaces like a CLI, HTTP APIs, or a WebAssembly module.

TRUE AGI MEANS 10% ECONOMIC GROWTH, SAYS SATYA NADELLA

Microsoft's CEO Satya Nadella says that true AGI should be measured by economic growth rather than AI advancements alone. The real benchmark for AGI's success, according to Nadella, should be global economic growth reaching 10%. The primary beneficiaries of AI advancements will be industries that use AI to drive productivity. The increased productivity will cause the economy to grow at a faster rate.

THE ODDS OF A CITY-KILLER ASTEROID IMPACT IN 2032 KEEP RISING. SHOULD WE BE WORRIED?

The odds of an asteroid, designated 2024 YR4, hitting Earth eight years from now continue to increase. Two weeks ago, the odds of the asteroid striking our planet were estimated at 1.9% - the most recent estimate has the likelihood at 3.2%. The asteroid is large enough to cause devastation near its impact site localized to a roughly city-sized area. This article contains an interview with Robin George Andrews, author of the recently published book How to Kill an Asteroid, where Andrews discusses what the impact odds mean and what humanity's plans are in the case of possible impact.

AI KILLED THE TECH INTERVIEW. NOW WHAT?

The nature of interviews will change drastically in the next few years - candidates will need senior engineering skills as soon as possible.

HOW APPLE'S CLOSED-SOURCE APPROACH IS LOSING OUT TO AI APP BUILDERS

40 of the top 100 iOS shopping apps are now non-native.

APPLE REVEALS C1, ITS FIRST IN-HOUSE 5G IPHONE MODEM, REPLACING QUALCOMM

The new C1 chip will debut in the iPhone 16e, the entry-model device that replaces the iPhone SE line.

MICROSOFT SHOWS PROGRESS TOWARD REAL-TIME AI-GENERATED GAME WORLDS

Microsoft Research's new World and Human Action Model (WHAM), can create low-resolution game prototypes.

GOOGLE'S NEW AI GENERATES HYPOTHESES FOR RESEARCHERS

Google's new multi-agent AI system, based on Gemini 2.0, is designed to help biomedical researchers discover new hypotheses and areas of research.

WHEN IMPERFECT SYSTEMS ARE GOOD, ACTUALLY: BLUESKY'S LOSSY TIMELINES

Knowing where it's okay to be imperfect allows developers to trade consistency for other desirable aspects and scale even higher.

more.