Best of AI — August 2025

  1. 1
    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.

  2. 2
    Article
    Avatar of lonely_programmerLonely Programmer·37w

    Slot Machine vs Vibe Coding

  3. 3
    Article
    Avatar of collectionsCollections·38w

    DeepSeek-V3.1 Release: A New Era in Open AI Technology

    DeepSeek has released V3.1, a 685-billion parameter open-source language model that rivals proprietary systems from OpenAI and Anthropic at significantly lower costs. The model features a hybrid architecture combining chat, reasoning, and coding capabilities, supports 128,000 token context, and achieves 71.6% on coding benchmarks. Available for free on Hugging Face with API compatibility, it's optimized for Chinese chips and represents a major step toward democratizing advanced AI technology.

  4. 4
    Article
    Avatar of aiAI·40w

    BREAKING: GitHub accidentally leaked GPT-5 details (proof inside)

    GitHub accidentally published and quickly deleted a changelog entry announcing GPT-5's general availability in GitHub Models. An archived version of the deleted page serves as evidence of the premature announcement, suggesting GPT-5 may be launching imminently.

  5. 5
    Article
    Avatar of controversycontroversy.dev·41w

    Enough is enough. Prompt engineering is not engineering.

    Argues that prompt engineering is fundamentally different from traditional software engineering, lacking the systematic design, mathematical rigor, and testable logic that define real engineering disciplines. The author contends that calling prompt writing 'engineering' is misleading marketing that inflates the perceived technical complexity of working with AI language models.

  6. 6
    Article
    Avatar of diamantaiDiamantAI·40w

    GPT-5 just proved something important - the scaling era is over

    The performance gap between GPT-4 and GPT-5 is smaller than previous generational leaps, signaling the end of the AI scaling era where bigger models automatically meant better performance. The future of AI development is shifting toward sophisticated engineering and AI agents built with existing models, rather than relying on massive compute budgets and larger model architectures.

  7. 7
    Article
    Avatar of dailydoseofdsDaily Dose of Data Science | Avi Chawla | Substack·39w

    8 RAG Architectures for AI Engineers

    Eight different RAG (Retrieval-Augmented Generation) architectures are explained with their specific use cases: Simple Vector RAG for basic semantic matching, Multi-modal RAG for cross-modal retrieval, HyDE for handling dissimilar queries, Self-RAG for validation against trusted sources, Graph RAG for structured relationships, Hybrid RAG combining vector and graph approaches, Adaptive RAG for dynamic query handling, and Agentic RAG for complex workflows with AI agents.

  8. 8
    Article
    Avatar of wheresyouredWhere's Your Ed At·37w

    AI Bubble 2027

    An MIT study reveals 95% of organizations get zero return from generative AI, while Meta freezes AI hiring and major outlets question if we're in a bubble. The analysis predicts the AI bubble will burst through a series of events over 18 months, including NVIDIA's growth slowing, AI funding drying up, major AI companies collapsing, and Big Tech pulling back from AI investments. Key vulnerabilities include OpenAI and Anthropic burning billions annually, CoreWeave's financial troubles, and AI startups raising at unsustainable valuations. The bubble is driven by vibes rather than returns, making it vulnerable to emotional market reactions when reality sets in.

  9. 9
    Article
    Avatar of vercelVercel·38w

    Introducing Streamdown: Open source Markdown for AI streaming

    Streamdown is a new open source Markdown renderer specifically designed for AI streaming applications. It handles unterminated chunks, interactive code blocks, and math expressions that traditional Markdown packages struggle with. The library includes Tailwind typography styles, GitHub Flavored Markdown support, Shiki syntax highlighting, LaTeX math rendering, and security features for untrusted content. It can be used standalone or as part of Vercel's AI Elements Response component.

  10. 10
    Article
    Avatar of hnHacker News·40w

    gpt-5 leaked system prompt

    A leaked system prompt reveals GPT-5's internal instructions and capabilities. The prompt shows personality guidelines emphasizing clarity and enthusiasm, memory management through a 'bio' tool, canvas functionality for document creation, image generation capabilities, Python code execution environment, and web search tools. It includes specific behavioral constraints like avoiding opt-in questions and copyright material reproduction.

  11. 11
    Article
    Avatar of theregisterThe Register·38w

    AWS CEO says AI replacing junior staff is 'dumbest idea'

    AWS CEO Matt Garman argues that replacing junior developers with AI is counterproductive, emphasizing that junior staff are cost-effective and naturally adept with AI tools. He warns that eliminating entry-level positions would create a knowledge gap in the future workforce. Garman advocates for using AI as a collaborative tool rather than a replacement, criticizing the metric of measuring AI value by lines of code generated. He recommends focusing education on critical thinking, problem decomposition, and learning skills rather than narrow technical abilities to prepare for rapid technological change.

  12. 12
    Article
    Avatar of webtoolsweeklyWeb Tools Weekly·38w

    CSS Tools, AI & LLMs, Git/CLI Tools

    A curated collection of developer tools across three main categories: CSS and HTML tools including pattern generators, hover effect creators, and 3D libraries; AI and LLM tools featuring coding agents, browser automation, and development assistants; and Git/CLI utilities for repository management, documentation generation, and terminal productivity. The newsletter also includes commercial tools and services for development infrastructure and productivity.

  13. 13
    Article
    Avatar of dockerDocker·39w

    GPT-5 Broke AI Apps: What Devs Must Do Now

    GPT-5's launch caused widespread AI application failures when OpenAI deprecated older APIs without warning. The incident highlights the brittleness of AI systems that depend on single providers or models. Modern AI applications are complex stacks involving prompts, embeddings, and retrieval logic that break when underlying models change. To build resilient AI systems, developers should implement AI High Availability (AIHA) architectures with multi-provider support, automated fallback mechanisms, behavioral monitoring, and contract testing. Key strategies include abstracting API layers, maintaining separate prompt libraries for different models, implementing graceful degradation, and treating model deprecation as an expected lifecycle event rather than an emergency.

  14. 14
    Video
    Avatar of hitenshowHiten Shah·40w

    The REAL reason Duolingo users are deleting the app

    Duolingo faced massive user backlash in May 2025 after CEO Luis von Ahn published a memo advocating for full AI automation with the phrase 'humans won't get us there.' While the company had been quietly integrating AI since 2020 and saw stock prices triple, users perceived a drop in content quality and felt the brand lost its personal touch. The crisis escalated when users called new AI-generated content 'AI slop,' leading to viral TikToks, Reddit criticism, and mass app deletions. Duolingo's response included deleting social media content and the CEO later walking back some statements, highlighting the challenge of balancing investor expectations with user experience.

  15. 15
    Article
    Avatar of devtoDEV·37w

    11+ V0 Alternatives That Every Developer Must Try

    A comprehensive comparison of 12+ AI-powered web app builders that serve as alternatives to V0 by Vercel. The tools range from production-ready code generators like Meku and Bolt.new to specialized platforms like Firebase Studio and Replit. Each platform offers different strengths: some focus on full-stack development with built-in authentication and payments, others prioritize quick prototyping or no-code solutions. The comparison covers key features, deployment options, code ownership, and target audiences for developers, founders, and non-technical users looking to build web applications using AI assistance.

  16. 16
    Article
    Avatar of zaidesantonManager.dev·39w

    The best time to be an Engineering Manager is now

    Engineering managers are positioned for success in an AI-driven future because their core value lies beyond coding. While AI agents may handle mid-level programming tasks, EMs excel in three critical areas: understanding business requirements and bridging technical-business gaps, defining clear technical solutions and requirements, and managing complex interpersonal relationships. These skills become more valuable as companies still need humans to guide AI agents, handle incidents, and navigate organizational dynamics. The key advice is for EMs to maintain current coding skills alongside their management expertise.

  17. 17
    Article
    Avatar of ergq3auoeReinier·38w

    Build & Deploy an AI-Powered Ecommerce Search Engine with Next.js 15, Hugging Face & Pinecone

    A comprehensive tutorial demonstrating how to build an AI-powered ecommerce search engine using Next.js 15, Pinecone vector database, and Hugging Face. The guide covers implementing semantic search capabilities that understand natural language queries instead of relying on exact keyword matching, creating a production-ready application with modern web technologies including TypeScript and Tailwind CSS.

  18. 18
    Article
    Avatar of arcjetArcjet·40w

    How long until we need to block Google?

    Google's AI Overviews are reducing website traffic as search results increasingly provide direct answers instead of linking out. Public companies report traffic declines from 52% to 28%, though Google claims overall click volume remains stable. Unlike OpenAI's granular bot controls, Google offers limited options for site owners to control how their content is used in AI features. Site owners can only block all Google crawling or allow everything, creating a dilemma as the traditional web traffic contract may be breaking down.

  19. 19
    Article
    Avatar of awsfundamentalsAWS Fundamentals·41w

    How to Generate AWS Architecture Diagrams with Amazon Q

    Amazon Q can automatically generate AWS architecture diagrams using MCP servers, eliminating the manual work of dragging boxes in traditional diagramming tools. The setup involves installing Amazon Q desktop app, Python packages for MCP servers, and Graphviz. While the tool can create basic diagrams quickly from plain English descriptions, it has limitations including non-deterministic outputs, inconsistent visual formatting, and difficulty achieving pixel-perfect results. Best practices include being specific with requirements, listing exact components and connections, and using structured prompts to avoid hallucinated components.

  20. 20
    Article
    Avatar of logrocketLogRocket·39w

    Using Grok 4 in the frontend development: Here’s what I’ve learned

    Grok 4 dominates academic benchmarks and mathematical reasoning but struggles with practical frontend development tasks. Testing shows it requires more iterations, costs significantly more ($3.31 vs $0.30-0.47 for competitors), and delivers inferior results compared to Claude Sonnet, Gemini, or Kimi K2 for UI builds, animations, and CSS work. While excellent for algorithmic challenges and backend-heavy features, Grok 4 ranks #12 in WebDev Arena compared to Claude's #1 position, making it an expensive choice for frontend developers.

  21. 21
    Video
    Avatar of fireshipFireship·40w

    GPT-5 is here... Can it win back programmers?

    OpenAI released GPT-5, claiming it's the first AI to outperform humans on certain benchmarks, but the reality is more nuanced. While GPT-5 unifies multiple models for better task routing and costs significantly less than competitors at $10 per million tokens, it still has limitations in coding tasks. Testing shows it can generate functional Svelte applications but makes errors with framework-specific rules. The model represents more of a consolidation effort than a revolutionary breakthrough, and programmers' jobs remain safe for now.

  22. 22
    Article
    Avatar of habrhabr·40w

    The Great Extinction: How AI is Destroying the Internet

    AI assistants are fundamentally reshaping the internet ecosystem by providing direct answers instead of directing users to websites, causing traffic declines for information sources like Wikipedia (23% drop) and Stack Overflow (64% drop). While entertainment platforms and social communities survive by offering emotional content and human interaction, the shift creates a monopoly on information access and threatens the diversity of knowledge sources. This creates a paradox where AI weakens the very sources it learned from, potentially leading to content degradation and the transformation of the internet from a global library into an amusement park dominated by a few AI giants.

  23. 23
    Article
    Avatar of dailydoseofdsDaily Dose of Data Science | Avi Chawla | Substack·38w

    JSON prompting for LLMs

    JSON prompting improves LLM outputs by providing structured format instead of vague natural language instructions. This technique leverages AI models' training on structured data from APIs and web applications, resulting in more consistent and predictable responses. JSON prompts eliminate ambiguity, enable output control, and create reusable templates for scalable AI workflows. While JSON is effective, alternatives like XML for Claude and Markdown also work well - the key is structure rather than specific syntax.

  24. 24
    Article
    Avatar of ergq3auoeReinier·38w

    Cursor Crash Course & AI Coding For Beginners

    A comprehensive guide covering the Cursor AI editor and AI-assisted coding workflows. The tutorial demonstrates key features including context management, coding rules, tab completion, and chat interface functionality to help developers integrate AI tools into their coding process.

  25. 25
    Article
    Avatar of pragmaticengineerThe Pragmatic Engineer·39w

    New trend: extreme hours at AI startups

    AI startups are increasingly adopting extreme work cultures requiring 80+ hour weeks, similar to China's "996" pattern. Companies like Cognition, Lovable, and xAI justify these demanding schedules as necessary to achieve AGI quickly before competitors. The promise of generational wealth through equity motivates employees to accept these conditions, as seen with Windsurf's acquisition by Google. However, long hours don't guarantee success, and this trend may persist due to intense competition and FOMO in the AI industry.

See all AI archives