Search the archive

SHOULD WE DECOMPOSE OUR MONOLITH?

The debate between adopting microservices, monolithic architectures, or a hybrid has evolved, with many companies now reconsidering their earlier migrations to microservices due to the complexities and overhead involved. This post outlines Theoretical Compliance Company's service architecture strategy, emphasizing the importance of balancing business growth with infrastructure efficiency while navigating the challenges of partial migrations and differing team priorities.

HOW TO SUCCEED IN MRBEAST PRODUCTION

A leaked onboarding document from MrBeast's production company reveals the unique culture and strategies behind creating viral YouTube content.

3 TOOLS TO MANAGE UP EASILY

Managing up is about aligning with your boss to create space for doing your best work with less oversight.

INTRODUCING NETFLIX'S KEY-VALUE DATA ABSTRACTION LAYER

Netflix's Key-Value Data Abstraction Layer (KV DAL) addresses challenges the company had with datastore misuse by providing a consistent interface layer over storage to application developers. This abstraction offers a two-level map architecture and supports usage through basic CRUD APIs, complex multi-item and multi-record mutations, and efficient handling of large blobs through chunking. It uses idempotency tokens, client-side compression, and adaptive pagination for predictable performance.

COMPILE AND RUN C IN JAVASCRIPT

Bun 1.1.28 introduces experimental support for compiling and running native C code directly from JavaScript. This allows developers to utilize system libraries and APIs that are not directly accessible through JavaScript, overcoming the limitations of N-API and WebAssembly. This article goes over how Bun does it by using TinyCC for fast in-memory compilation that provides near-zero call overhead.

HOW TO REASON ABOUT EXCEPTIONS

While every function has the potential to throw an exception, it's more helpful to think about whether the entire function succeeds or fails rather than focusing on individual operations that could cause errors. Make functions more simple to improve readability and reasoning and write clean-up code that keeps a consistent program state regardless of how the function exits.

STOP DESIGNING YOUR WEB APPLICATION FOR MILLIONS OF USERS WHEN YOU DON'T EVEN HAVE 100

It doesn't make sense to over-engineer web apps by focusing on scalability from the beginning, especially when the user base is still small. Spending time and resources on building for millions of users before even having 100 users is a waste of time and money. It is more important to focus on creating a product that people want, gathering user feedback, and iterating quickly.

STOP USING SERIAL IN POSTGRES

It's better to use identity columns instead of using the serial data type in Postgres. This is because `serial` has several issues, like permission complexities, a lack of integrity guarantees, and awkward ergonomics. Identity columns provide a better way to manage auto-incrementing primary keys and are also compliant with the SQL standard.

🛝 TRY THE QRVEY DEVELOPER PLAYGROUND (UNGATED)

TS-REMOVE-UNUSED

This tool automatically cleans up TypeScript code by removing unused exports, deleting modules with no referenced exports, and reporting these changes without making modifications in check mode.

CONFORM

Conform is a type-safe form validation library that uses web fundamentals to progressively enhance HTML Forms. It gives users control of the full form control lifecycle and has full support for server frameworks like Remix and Next.js.

SELECTABLE

Selectable is a mobile app that lets you access your Postgres databases with a mobile-optimized experience. It lets you write complex SQL queries easily on the go.

TECH JOBS ARE GONE AND NOT COMING BACK

The tech job market has changed a lot, with constant layoffs and hiring freezes. The surge in hiring during the pandemic has reversed, leading to a decrease in demand for software developers and a struggle for entry-level positions. While companies are focusing on AI development, leading to high demand for specialists in that field, the overall job market for tech workers is much more competitive than before.

LEGACY MODERNIZATION MEETS GENAI

Generative AI can be used to modernize legacy code. This article goes over a tool that combines an LLM with a knowledge graph derived from a legacy codebase, allowing it to understand existing code and extract low-level requirements, build high-level explanations, and even generate automated documentation.

PAYING DOWN TECH DEBT: FURTHER LEARNINGS

While it's important to address tech debt incrementally, sometimes a codebase requires a significant rewrite. Rewrites like these need management support, careful planning, and stakeholder buy-in.

AI-GENERATED CODE IS CAUSING OUTAGES AND SECURITY ISSUES IN BUSINESSES

AI-generated code is causing downtime and security issues in businesses due to insufficient code reviews and developers' tendency to trust AI too readily.

HOW MUCH GPU MEMORY IS NEEDED TO SERVE A LARGE LANGUAGE MODEL (LLM)?

A popular LLM interview question is to calculate the GPU memory needed to run inference for different LLMs - this can be done with a simple formula.

SIMPLE ICONS

A collection of over 3,000 free SVG icons for popular brands.

THE ARRIVAL OF JAVA 23

Java 23 is now available, with language improvements, API updates, performance optimizations, and tooling changes.

Schedule a demo to learn more

YOUTUBE HYPE GIVES SMALLER CREATORS A PLACE TO SHINE

YouTube's new Hype feature is aimed at growing smaller channels and helping people discover and share new creators. It introduces a new button for hyping a video - the most hyped videos will appear on a platform-wide leaderboard. Leaderboards are specific to each country. YouTube wants to make sure the leaderboard changes a lot and that people feel invested in each thing they type. Users will receive a summary of the videos they hyped and how well they performed every week.

HERE'S HOW GREEN BUBBLES ARE GETTING UPGRADED IN IOS

iOS 18 introduces RCS support for messaging, enabling chats between iPhone and Android users to have many features core to other messaging apps. Cross-platform chats will now support high-resolution media, typing indicators, read receipts, and more reliable group messaging. Not all carriers support RCS on iPhone yet. Apple's RCS support doesn't include cross-platform end-to-encryption between iOS and Android.

WORLD-FIRST ADHD HEADBAND TREATS SYMPTOMS IN 20 MINUTES PER DAY

Neurode is an Australian startup with a device that treats ADHD using non-invasive light electrical brain stimulation delivered by a headband worn for just 20 minutes a day. It helps manage focus, memory, and impulse control and can measure brain activity and track changes over time through an app. The headband and app feature an AI assistant to help users reduce overwhelm and help form personalized to-do lists. It is unclear how much it will cost or how effective it is, but those in the US who are curious can now sign up to beta-testing the headband.

HUMANOID ROBOTS HEAD TO HOMES: MEET THE NEO BETA

studies.

REACTR

The React ecosystem is rich with components that can enhance R web and Shina apps. reactR provides a set of convenience functions for using React in R. It aims to allow R users to more easily incorporate React and JSX. reactR can be used to create rich visualizations and highly customizable charts from R data.

THINGS I'VE LEARNED SERVING ON THE BOARD OF THE PYTHON SOFTWARE FOUNDATION

The Python Software Foundation (PSF) is a non-profit organization with a mission to promote, protect, and advance the Python programming language and to support and facilitate the growth of a diverse and international community of Python programmers. It is responsible for protecting and supporting the Python language and community, but not the development of CPython itself, which is handled by the Python core team, who are governed by the Python Steering Council. This article looks at how the PSF functions, what life is like as a board member, the kinds of things the board talks about, and more.

WHAT AMAZON'S STRICT RETURN TO OFFICE POLICY SAYS ABOUT REMOTE WORK

Amazon's move to push employees back to the office five days a week signals that tech sector employers have regained some leverage over workers. The tech sector has experienced a lot of layoffs over the past few years so it's not easy to find a new job, leaving many feeling stuck. Amazon's change may give other companies cover to restrict work-from-home a bit more. However, research shows that most companies have settled into a new hybrid setup and are not likely to go back.

IN US V. GOOGLE, YOUTUBE'S CEO DEFENDS THE GOOGLE WAY

The central question of US v. Google is whether Google bought its competitors in the online advertising business only to park them so no one could possibly upset its dominance. YouTube's CEO Neal Mohan testified that Google was never attempting to buy up and neuter its competitors and that it was simply trying to compete. Google has previously purchased competitors at above market valuation to integrate their technology. The company claims the acquisitions were only in service of building better products in a competitive space.

and tips to write a winning welcome email. Download for free

A FRIENDLY INTRODUCTION TO ASSEMBLY FOR HIGH-LEVEL PROGRAMMERS

This is an introduction to assembly for people with programming experience - it discusses the basics of the language and maps them to high-level programming constructs.

GM ELECTRIC VEHICLES CAN FINALLY ACCESS TESLA SUPERCHARGERS

Owners of GM electric vehicles will need to purchase 'GM approved' Tesla adapters through each brand's smartphone app for $225 - future GM vehicles will come with Tesla's charging port natively installed.

VALKEY 8.0

Valkey 8.0 includes stability and performance improvements over the second release candidate and is fully compatible with Redis OSS 7.2.4.

READ EASIER WITH OUR FREE FONT

Atkinson Hyperlegible was designed to help low-vision readers by improving legibility and readability through the use of clear and distinctive letters and numbers.

KUBESAFE

Kubesafe allows developers to safely run commands across multiple Kubernetes contexts.

GOODHART'S LAW IN SOFTWARE ENGINEERING

Goodhart's Law - when a measure becomes a target, it ceases to be a good measure - is a warning for bosses who use terrible metrics.

Add WorkOS to your app in minutes ↗️

SALESFORCE'S NEW AI STRATEGY ACKNOWLEDGES THAT AI WILL TAKE JOBS

Salesforce is overhauling its AI strategy, introducing generative AI tools for tasks without human supervision and revising its pricing model to $2 per AI-driven conversation. This shift aims to address investor concerns about AI-induced job losses impacting subscription-based revenues. The new AI tools offer higher efficiency and autonomy compared to traditional copilots and chatbots.

QWEN 2.5

An impressive array of open models that approach the frontier of performance. Specifically, they have strong performance on code, math, structured output, and reasoning. The Qwen team has also released a suite of sizes for a variety of use cases.

CREATE FULL WEB APPS WITH LLAMACODER

Together AI and Meta have partnered to make a tool that enables users to build entire apps from just a prompt on the LlamaCoder platform. It is akin to Claude Artifacts but purely created to show off the speed of Together AI's inference engine.

BITNET WITHOUT RETRAINING

The Hugging Face team, using a new technique called quantization warm up, fine-tuned Llama 3 8B and maintained Llama 1 performance while quantizing it to use 1.58 Bits per parameter.

1X WORLD MODEL

1x, a robotics company, has trained a video generation model that can be used to simulate first person views of robotic activity. This can be helpful for offline data generation and robot training.

DIAGRAM OF THOUGHT: ITERATIVE REASONING IN LANGUAGE MODELS

The Diagram of Thought (DoT) framework introduces a new way for large language models to reason by organizing ideas into a directed acyclic graph (DAG). This method allows models to propose, critique, refine, and verify ideas, improving logical consistency.

ZML INFERENCE

ZML is a lightning fast inference engine written in Zig that is designed with performance in mind. It can be used for a variety of models, although it requires some customization to get working for a new architecture.