Search the archive

RUBY'S NEW TLDR TESTING FRAMEWORK

HOW TO WRITE BETTER DJANGO CODE

STRONG STATIC TYPING, A HILL I'M WILLING TO DIE ON...

OPENTOFU

OpenTofu is an open-source tool for building and managing infrastructure safely and efficiently. It can handle both popular service providers as well as custom in-house solutions.

SHUTTLE

Shuttle is a Rust-native cloud development platform that lets you deploy your Rust apps for free. It comes with features that improve developer productivity. It has first-class support for popular Rust frameworks, automatic resource provisioning, and scalable hosting.

CSS SCOPE INLINE

CSS Scope Inline allows web developers to scope inline style tags in pure vanilla CSS. It has no dependencies. CSS Scope Inline is an alternative to Tailwind CSS.

SECURITY WEAKNESSES OF COPILOT GENERATED CODE IN GITHUB

A recent study found that 35.8% of Github Copilot-generated code snippets contain security weaknesses. The most frequent security issues include OS Command Injection, Use of Insufficiently Random Values, and Improper Handling of Exceptional Conditions, with 11 of them being part of the 2022 Common Weakness Enumeration (CWE) Top-25 list. The study emphasizes the importance of cautious code review and security checks when using AI-generated code in software projects.

DEVELOPER?

GENERATIVE AI

KATA CONTAINERS

PICKING A VECTOR DATABASE: A COMPARISON AND GUIDE FOR 2023

FROM RAPID HIRING

GMAIL IS GETTING EMOJI REACTIONS 🤩

Google is bringing emoji reactions to its Gmail service. It will roll out gradually starting with Android users and to other platforms over the next few months. Users with access will see a smiling emoji at the bottom of messages. Some emojis will enable full-page animations. Emoji reactions will only show up in the Gmail app - people using third-party email clients will receive emojis as separate emails. The feature is not available to users on school or work accounts or for large group messages.

AMAZON REPORTEDLY USED A SECRET ALGORITHM TO JACK UP PRICES

Amazon deployed a secret algorithm called Project Nessie to gauge how high it could raise prices before competitors stopped increasing their prices as well. The algorithm would inflate prices and monitor whether other retailers would follow suit. If retailers maintained the lower price, the algorithm would revert back to the normal price. Amazon reportedly stopped using it in 2019. Project Nessie is just one of the many ways the FTC has accused Amazon of illegally maintaining its market dominance in the e-commerce industry.

STARLINK LAUNCHES V2 MINI-SATELLITES WITH 'SPACE LASERS'

Starlink has launched new versions of its satellites that can communicate via lasers. The service has an estimated one million subscribers and is looking to improve its offering as competition warms up in the sector. While Amazon's Kuiper project has so far failed to take off, another rival, OneWeb, is expected to begin offering services soon. Starlink's new satellites will be able to transfer data through laser links at speeds of 100 Gbps over thousands of miles.

THE WORLD’S FIRST ARTIFICIAL ENERGY ISLAND JUST GOT THE GO-AHEAD

Princess Elisabeth Island, the world's first artificial energy island, has secured its environmental permit in Belgium. The island is an at-sea electricity grid that will connect offshore wind farms to the Belgian mainland. It will serve as a hub for future interconnectors with the UK and Denmark. Construction is planned to be completed in August 2026. The island is expected to be fully connected to all wind farms and the mainland by 2030. A 3-minute video about the project is available.

POSTGRES

Database schema migrations are essential for keeping systems up to date and in sync with evolving application requirements, but they come with many challenges. Many developers choose to avoid complex migrations due to this and only make additive changes, resulting in an accumulation of technical debt in the database schema. pgroll is a command-line tool for performing schema migrations against Postgres databases. It allows users to define schema migrations using a high-level JSON format and then takes care of executing them.

ZANY

zany is an easy-to-use, configurable favicon server. Developers can simply point their icon tag at zany and configure styling without generating, downloading, or serving favicons. It supports text, emojis, background color, text color, and font-family. zany favicons can be easily tested in the browser bar.

ELON'S DECISION MAKING: AN ANECDOTE COMPILATION

Elon Musk questions life's default settings and goes all-in into his work. If one has enough awareness when faced with a new problem, it's easier to come up with novel ways to solve it. This article contains a compilation of anecdotes from Walter Isaacson's biography on Elon Musk of times when Musk made unusual strategic or tactical decisions and what the outcome was. Musk's willingness to have difficult conversations is something many can learn from.

THE ONLINE MAZE OF JOB APPLICATIONS

Applying to jobs online is like navigating a maze. Resume parsing software, application tracking systems, and other factors make the process pretty frustrating. This article explores factors that make job applications more or less frustrating. The author applied to 250 jobs manually to see how long it took to get from start to finish. They applied for engineering/product-focused roles in a mix of companies across different industries.

WHAT TO EXPECT FROM SBF'S TRIAL — AND AFTER

While the proceedings may leave Sam Bankman-Fried feeling very bleak about his future, they will hopefully help make customers whole, hold bad actors accountable, and restore the public's trust in markets.

HONEY, I SHRUNK THE NPM PACKAGE

This article looks at the realm of compression and explores the possibilities of modernizing npm's compression strategy.

WHAT IT'S LIKE TO START A JOB AS CEO

You have to learn a lot very fast, make an effort to connect with people, make big decisions quickly, and more, when first onboarding as a CEO.

FUTURE OF AI ART

This interview with Microsoft CTO Kevin Scott covers Microsoft's AI efforts, hardware infrastructure, the future of AI art, and more.

FASTSTREAM

FastStream is a Python framework for building asynchronous services that interact with event streams.

THE 3 WAYS TO BALANCE MONEY AND MEANING

The three archetypal ways people approach balancing money and meaning are the deferred life plan, being bi-vocational, or choosing to integrate.

REKA LAUNCHES THEIR FLAGSHIP MULTIMODAL MODELS

Yasa-1 is an audio, video, and text model from the creators of the extremely popular T5 language model. It boasts impressive performance across a number of tasks. Yasa-1 is currently in private preview.

LINKEDIN GOES BIG ON NEW AI TOOLS

LinkedIn is launching a series of AI-based features to enhance its job hunting, marketing, and sales offerings. These include an improved version of its Recruiter talent sourcing platform with integrated AI, an AI-driven LinkedIn Learning coach, and a new AI tool for marketing campaigns.

Despite open-source AI democratization, over 60% of generative AI job postings from July 2022-2023 were in 10 major tech hubs. Interventions and expanded research programs are needed to diversify AI's geographic footprint.

GENERALIST ROBOT POLICY

A robot's policy is the guiding algorithm that determines how it acts. Typically these are robot and task specific. This makes it hard for robotics to adapt. This new work from Google shows a single (55B parameter) policy that can perform 160k+ different tasks.

MAKING MULTI-MODAL LLMS WORK FOR DIFFERENT TASKS

Multi-modal Large Language Models excel in many tasks but often struggle to pinpoint specific objects in images. Researchers have developed a way to improve this by turning object locations into text and training the model with special tasks.

EFFICIENT LARGE-SCALE 3D SCENE RECONSTRUCTION

Rebuilding large 3D scenes, especially when some data is missing, is crucial for self-driving cars. This new method, PC-NeRF, combines two modules to optimize the reconstruction at multiple levels, ensuring detailed 3D models even with incomplete data.

BETTER IMAGE RESTORATION WITH VISION-LANGUAGE MODELS

Vision-language models like CLIP work well for many tasks but struggle with image restoration. This project introduces DA-CLIP, a model that adjusts how CLIP works to handle these tasks better, leading to clearer image reconstructions.

ONNXSTREAM

A program that enables you to run Stable Diffusion 1.5 on a Raspberry Pi Zero 2.

LARGE WORLD MODELS FOR SELF DRIVING

Most self driving systems have been large combinations of rules-based, expert systems, and learned modules. This work by Wayve shows a 9B joint video, action, and text model that performs extremely well in diverse scenarios.

IS DATA STILL A MOAT?

Generative AI is reshaping the traditional value placed on vast data collections. As large language models can be fine-tuned with minimal data and even generate synthetic datasets, the uniqueness and importance of proprietary data may be waning.

AI IS GETTING BETTER AT HURRICANE FORECASTING

Hurricane Lee's path was predicted by AI-based models from Nvidia, Huawei, and DeepMind. These AI models, which analyze patterns from historical atmospheric data, have shown comparable or sometimes better performance in storm tracking than conventional models, delivering rapid results in minutes instead of hours, and might play a significant role in official forecasts in the coming years.

THE NATIONAL SECURITY AGENCY IS STARTING AN AI SECURITY CENTER

The National Security Agency is starting an artificial intelligence security center in order to keep pace with the rapid integration of AI into U.S. defense and intelligence systems.

SEC LOSES AGAIN—APPEAL OF LANDMARK RIPPLE RULING REJECTED

A US federal judge rejected the SEC appeal against a judgment that favored crypto firm Ripple Labs. According to the judge, the SEC failed to prove that there was a problem with the initial decision in the case. For context, the court had previously ruled against the categorization of Ripple's cryptocurrency token sale as an illegally unregistered securities offering. The SEC can still appeal following the case's formal conclusion next April.

CELSIUS LAWYER SAYS REVIVAL PLANS INCLUDE $450M IN SEED FUNDING

Bankrupt crypto lender Celsius has laid out revised plans to exit bankruptcy with $450 million in fresh funding from Fahrenheit LLC to kick off the restart. The relaunched company could then work as a "user-owned Bitcoin miner", aiming to begin repayments to users by the end of 2023. The recovery plan still needs approval from securities regulators and the bankruptcy judge. It could face opposition from some creditors who believe the new firm is overvalued.

HELIUM HOTSPOTS GO LIVE IN BID TO TURN MIAMI NETWORK PROFITABLE

Nova Labs, the team behind the Helium wireless network, is releasing at-home hotspots in Miami with a price tag of $250. The new hotspot rigs are part of Helium's strategy to increase data usage on Helium hotspots in the place of regular cellular data usage from providers like T-Mobile.

THE OP STACK’S FAULT PROOF SYSTEM IS LIVE ON OP GOERLI

The OP Stack has launched its first fault proof system on the OP Goerli Testnet. The system is made up of a Fault Proof Program, a Fault Proof Virtual Machine, and a dispute game protocol, all designed to secure the network and ensure trust. The hope is that a robust fault proof system will enhance security across all OP Chains and OP Stack chains.

5 LESSONS FROM CRYPTO FUND INVESTING

Longtime crypto Investor Ben Jacobs reveals his five main lessons from crypto fund investing. He emphasizes that managing risk is crucial due to liquid tokens, warns about alpha decay due to crypto's relatively small market cap, and points out that many funds fail. He highlights that fund structure plays a key role and ends on a positive note indicating that the possible returns in the current climate have never been better.

LOOKING INTO ARBITRUM GRANT PROPOSALS

This post goes over Arbitrum grant proposals and discusses the merits of 7 different projects that applied. Four categories of grants were listed, with each category having different requirements. These applications came on the heels of Arbitrum’s announcement that it would be funding projects in the ecosystem with 50 million ARB.

WHY FRIEND TECH IS GREAT (MAYBE THE BEST EVER PONZI)

This blog post discusses FriendTech and describes it as a platform that seems to transform the concept of social signaling by showcasing personal value and reputation in a quantifiable manner. Users are encouraged to engage in genuine positive behavior and discourage fraud, with implications of a reputational game. Overall, FriendTech could become a significant social layer and leaderboard, evolving into a protocol for online reputation management.