Best of AI — 2025

  1. 1
    Article
    Avatar of antonzAnton Zhiyanov·33w

    Write the damn code

    Software engineers should avoid endless prompt refinement when working with AI and instead actively write code themselves. Rather than trying to perfect prompts to get AI to generate complete solutions, developers should engage with the code directly by refactoring AI-generated code, writing initial versions for AI review, handling critical parts manually, or creating code outlines for AI to complete. This hands-on approach produces better results than attempting to 'program in English' through prompt iteration.

  2. 2
    Article
    Avatar of workchroniclesWork Chronicles·1y

    (comic) AI won't take everyone's job

    A comic piece illustrating that AI will not take over everyone's job, easing concerns about job security in the age of automation and emphasizing the importance of workplace culture.

  3. 3
    Article
    Avatar of javarevisitedJavarevisited·1y

    10 Things Software Engineers Should Learn in 2025

    In 2025, software engineers should focus on mastering skills like system design, cloud computing, machine learning, artificial intelligence, generative AI, DevOps, technical writing, app development, cybersecurity, and data engineering. Resources such as online courses and certifications can aid in learning these crucial topics, ensuring readiness for the evolving tech landscape.

  4. 4
    Article
    Avatar of hnHacker News·30w

    Every Fucking AI-Coded Website Ever

    A satirical critique of AI-assisted coding practices, highlighting common pitfalls like blindly copying generated code without understanding it, poor project organization, lack of testing and documentation, and security vulnerabilities. The piece mocks developers who rely entirely on AI tools like Claude and ChatGPT to build websites without learning the underlying technology, resulting in identical-looking sites with messy codebases and questionable quality.

  5. 5
    Video
    Avatar of fireshipFireship·1y

    OpenAI o3 tries to curb stomp DeepSeek...

    Recent restrictions have seen the banning of Deep Seek by countries like Italy, the US, Australia, and Taiwan. Meanwhile, OpenAI has launched the new 03 Mini model and a Deep Research feature for Pro users, aiming to remain competitive. These developments are part of a broader trend in the AI landscape, with open-source solutions making rapid progress. Despite corporate efforts, some AI tools face performance issues, and Google's Gemini has similar features to OpenAI's new offerings.

  6. 6
    Article
    Avatar of lobstersLobsters·1y

    Why do AI company logos look like buttholes?

    AI company logos frequently feature circular designs with gradients, central openings, and radiating elements, often leading to unintended anatomical comparisons. Factors like design psychology, unintentional biomimicry, the copycat effect, and design by committee contribute to this trend. This phenomenon highlights the tech industry's fear of standing out, with historical design trends showing similar patterns of conformity.

  7. 7
    Article
    Avatar of cassidooCassidy's blog·22w

    Vibe coding is boring

    Vibe coding with AI agents is effective for shipping side projects quickly, but removes the satisfaction and learning that comes from hands-on development. While tools like GitHub Copilot and Spec Kit can automate implementation from specifications, watching agents write code is tedious and lacks the joy of problem-solving. The author reserves AI-assisted coding for projects where only the final output matters, preferring to manually build applications where the tech stack or implementation details are interesting.

  8. 8
    Article
    Avatar of devblogsDevBlogs·1y

    A 10x Faster TypeScript

    TypeScript is undergoing a major performance improvement by porting its compiler and tools to a native implementation. This effort aims to improve editor startup, reduce build times by 10x, and decrease memory usage. Initial previews will be available by mid-2025, with a full feature release by the end of the year. The native port promises significant speed gains in editor responsiveness, command-line builds, and will support more advanced development features.

  9. 9
    Article
    Avatar of daily_updatesdaily.dev Changelog·49w

    daily.dev now powered by Claude Sonnet 4!

    Daily.dev has upgraded its content pipeline to use Claude Sonnet 4, bringing improvements to TLDR summaries, content tagging, recommendations, and content filtering. The AI model's enhanced language comprehension and contextual awareness aims to deliver more relevant and accurate content curation for developers on the platform.

  10. 10
    Article
    Avatar of gopenaiGoPenAI·1y

    How to Build a Local RAG with DeepSeek-R1, LangChain, and Ollama (Step-by-Step Guide)

    Learn how to build a local Retrieval-Augmented Generation (RAG) system using DeepSeek-R1, LangChain, and Ollama. This guide details the installation, setup, and deployment of a RAG pipeline that processes PDFs locally, ensuring data privacy, cost efficiency, and customizability. The solution utilizes ChromaDB for document retrieval and Streamlit for a user-friendly interface.

  11. 11
    Video
    Avatar of thecodingslothThe Coding Sloth·1y

    How To Program Apps That Make INFINITE MONEY

    Learn how to create and market apps that can generate income. The guide covers finding an idea, building the app, marketing strategies, and useful tools like AI code editors and design platforms. The key focus is on practicality and speed rather than perfection, highlighting the importance of SLC (Simple, Lovable, Complete) in app development.

  12. 12
    Article
    Avatar of devtoDEV·41w

    Programming Is Becoming Prompting

    The programming landscape is shifting as AI tools transform coding from writing functions to crafting prompts. While AI assistance can scaffold codebases, generate tests, and speed up routine tasks, it risks diminishing creativity and problem-solving skills. Developers need to balance leveraging AI for efficiency while maintaining deep coding knowledge for debugging, scaling, and handling complex edge cases. The key is knowing when to use AI and when to code manually, as understanding fundamentals remains crucial when AI-generated solutions break or need customization.

  13. 13
    Article
    Avatar of devsquadDev Squad·1y

    🚀 14 Must-Bookmark Developer Resources to Boost Your Productivity

    Discover 14 essential resources that can significantly enhance your productivity as a developer. These include interactive coding exercises, free learning platforms, AI-powered development assistants, design inspiration tools, backend and API services, deployment solutions, and resources for job hunting and resume optimization.

  14. 14
    Article
    Avatar of uxplanetUX Planet·1y

    UI Design with ChatGPT 4o

    ChatGPT 4o can now generate sophisticated UI designs with real text by using detailed prompts. Although it may not produce an exact pixel-perfect design, it is useful for quickly experimenting with different ideas. The post provides a step-by-step guide on how to use ChatGPT for UI design, the importance of detailed prompts, and recommendations for refining designs using tools like Figma.

  15. 15
    Article
    Avatar of awegoAwesome Go·1y

    Cursor for Large Projects

    Cursor and Claude are effective not just for prototyping but also for maintaining large software projects. Their use can accelerate development processes by structuring, refactoring, and testing code efficiently. Stream CEO Thierry shares a workflow for using Cursor to manage a large Go codebase, emphasizing the importance of a good edit and test loop and the use of dedicated documentation folders for AI. The guide also covers steps for setting up Cursor, highlights the combination with Goland for debugging and refactoring, and provides tips for maximizing AI benefits in software development.

  16. 16
    Article
    Avatar of dailydoseofdsDaily Dose of Data Science | Avi Chawla | Substack·1y

    10 MCP, AI Agents, and RAG projects for AI Engineers

    Explore 10 AI-focused projects including building an MCP-powered Agentic RAG, a multi-agent book writer, and a RAG system that understands audio content. Learn how to build and fine-tune AI models like DeepSeek-R1 and create applications using open-source tools like Llama 4 and Colpali.

  17. 17
    Article
    Avatar of iotechhubiO tech_hub·1y

    What is WebLLM

    WebLLM, developed by the MLC-AI team, allows large language models (LLMs) to run fully within a web browser using modern web technologies like WebAssembly and WebGPU. This enables models to be more accessible client-side, providing privacy and offline support. While cloud-based LLMs are faster and require powerful servers, WebLLM offers cross-platform portability and easier installation. Implementation can be done using the WebLLM npm package, which includes support for web workers to enhance application performance.

  18. 18
    Article
    Avatar of hnHacker News·31w

    I'm in Vibe Code Hell

    The coding education landscape has shifted from "tutorial hell" (passive video consumption) to "vibe code hell" where learners over-rely on AI coding assistants. While students can now build projects faster with tools like Cursor and Claude, they often fail to develop deep understanding of how software works. AI tools can be valuable for learning when used as Socratic tutors rather than code generators, but the sycophantic nature of LLMs and their tendency to agree with users creates new learning challenges. Effective learning still requires discomfort and independent problem-solving, whether that means turning off tutorial videos or disabling AI autocomplete.

  19. 19
    Article
    Avatar of pragmaticengineerThe Pragmatic Engineer·1y

    Stack overflow is almost dead

    The engagement on Stack Overflow has significantly decreased, with the number of questions asked monthly hitting levels from its early days in 2009. Key factors contributing to its decline include enhanced moderation efficiency reducing question flow starting in 2014, the impact of ChatGPT offering quick and polite answers trained on Stack Overflow data since November 2022, and outdated moderation policies. Additionally, the site was acquired by Prosus in 2021, which preceded a marked fall in activity.

  20. 20
    Video
    Avatar of mattpocockMatt Pocock·1y

    Zod 4 is CRAZY

    Zod 4, a popular schema validation library, introduces significant improvements including enhanced performance in TypeScript, faster string, array, and object parsing, and reduced bundle size. New features include Zod Mini for front-end optimization, JSON schema conversion, a schema registry, and an interface for defining object types. Additionally, it provides validation for file instances, a global translation API for error messages, and prettier error printing.

  21. 21
    Video
    Avatar of youtubeYouTube·1y

    Build Everything with AI Agents: Here's How

    David Andre demonstrates how to build AI agents even for beginners using n8n, a no-code automation tool. He details the process of setting up triggers, integrating Telegram, and handling both text and voice messages. By adding tools such as Gmail and Google Calendar, he shows how to create powerful AI agents capable of automating various tasks. He also highlights the value of continuous testing and the potential productivity boosts these agents can provide.

  22. 22
    Article
    Avatar of hnHacker News·1y

    Get the hell out of the LLM as soon as possible

    Large Language Models (LLMs) should not be used for decision-making or implementing business logic due to their poor performance in these areas. Instead, LLMs should be employed as an interface for translating user inputs into API calls, with the actual logic handled by specialized systems. This approach enhances performance, debugging, and reliability. LLMs are best utilized for tasks involving transformation, interpretation, and communication, rather than maintaining critical application state.

  23. 23
    Article
    Avatar of theregisterThe Register·25w

    Linus Torvalds: Vibe coding is fine, but not for production

    Linus Torvalds shares his perspective on AI-assisted coding, stating that while vibe coding can help newcomers get started with programming, it's unsuitable for production code due to maintenance concerns. He discusses Rust's gradual integration into the Linux kernel, noting it has taken longer than expected but is becoming a real part of the codebase. Torvalds addresses AI's impact on kernel development, mentioning issues with crawlers disrupting infrastructure and AI-generated bug reports, though these problems are less severe than in other projects. He compares AI to previous productivity tools like compilers, suggesting it won't eliminate programming jobs but will change how developers work.

  24. 24
    Article
    Avatar of bytebytegoByteByteGo·1y

    EP158: How to Learn API Development

    Key tips for learning API development including fundamentals, request/response handling, authentication and security, and deployment. A focus on AI coding aids like Augment Agent, and significant coding tools for 2025. Also covers network protocol dependencies and essential design patterns. ByteByteGo is hiring for a sales and partnerships lead.

  25. 25
    Article
    Avatar of javarevisitedJavarevisited·1y

    5 Best Books to Learn AI and LLM Engineering in 2025 (That Aren’t a Waste of Time)

    Discover the top five books recommended for mastering AI and LLM engineering in 2025. These selections focus on practical systems design, deployment, and real-world applications, helping readers save time and effectively build production-ready models. Written by experienced practitioners, these books offer guidance for those serious about becoming proficient in large language models and AI systems.

See all AI archives