标签：#ai

YouTube Shorts is getting a makeover.

I spent three iterations on an auto-fix pipeline that still doesn't work reliably. Here's what I learned. Loop 1 Wrote a background script. Pull tickets from Azure DevOps, run them through a local model, hand to a coding agent, push the result. Poll → triage → fix → push. Worked 40% of the time on trivial tickets. Anything that crossed file boundaries or needed real context — stalled or hallucinated. I shipped it anyway. That was naive. Loop 2 Made it smarter. Pre-selected relevant files. Broke big tickets into subtasks. Turned complex edits into atomic steps with verification between each. Got it to 55% or so. But every fix created two new edge cases. The complexity was compounding faster than the reliability. Loop 3 Went all in. Embeddings for dedup. Multi-repo routing. Auto-revert. A learning loop that fed failures back into future runs. The model server started dying. 890 memory errors in a day. Root cause: two independent consumers hitting the same local model server, each with its own retry loop. When memory filled up, retries amplified instead of staggering. The system was making itself worse. Fixes were simple in hindsight — stop retrying OOM, serialize access, use the local binary not npx. But the pattern kept repeating: add more to fix the last thing, break something else. Where I'm At The pipeline still only works on easy tickets. Hard ones need a human. After three rounds, the main thing I learned is that local models hit a wall before your ambition does — not in quality, in working memory. And adding features doesn't fix reliability gaps. It just moves them around. The 507 retry spiral taught me more than any successful deploy this year. Because it was entirely my fault. Not the model's, not the framework's. I built concurrent consumers with independent retry loops and expected them to coordinate. They didn't. What's Next I'll do a fourth loop. Smaller. A dedicated fast model for cheap work, the big model only for editing. One consumer at a time. Might

After months of building in public, is live on Product Hunt today! 🎉 The problem Every dev team I've worked with ends up in the same trap: Jira is too heavy and slows everyone down with ceremony, while Trello/Notion are too light and can't actually tell you if your sprint is on track. Nobody had a tool that combined real project management with AI that actually understands developer workflows. So I built one. What is Rahnuma.io? Rahnuma.io is an AI-powered DevOps platform that sits where project management meets your actual engineering workflow: 🧠 AI task generation — describe a feature in plain English, get a structured task with subtasks 📊 Deadline risk forecasting — a live risk score (0–100) built from time risk, blockers, and completion rate, so you know a sprint is in trouble before it blows up 🗂️ Kanban + sprints — drag-and-drop boards, WIP limits, burndown charts, story points 🤖 AI sprint retros — auto-generated "what went well / what didn't / action items" 🔗 GitHub & Bitbucket integration — see commits next to the tasks they belong to 📈 Reports that don't require a translator — including an "Explain to My Boss" button that turns your sprint into a plain-English executive summary 🔔 Slack notifications for task creation, assignment, and comments 🧾 Client portals — shareable, printable progress reports for non-technical stakeholders Why it's different Most "AI project management" tools just bolt a chatbot onto a Trello clone. AI is wired directly into the data model — it knows your sprint history, your velocity, your blockers — so the forecasts and summaries are grounded in your actual project state, not a generic prompt. Tech under the hood Next.js 15 (Turbopack) · TypeScript · Prisma + PostgreSQL · Clerk auth · Polar billing · Redis · xAI Grok with Groq fallback for AI · Server-Sent Events for realtime sync. Try it It's free to start, no credit card required. I'd love your feedback — especially from anyone who's felt the Jira-is-too-much / Trello-is-too-littl

NSFW uses account for "well over half" of traffic, a new report says.

The Trump administration, apprehensive of potential security issues, has reportedly asked OpenAI to stagger the release of its next big-ticket model, GPT-5.6. The Information reported that OpenAI CEO Sam Altman told employees Wednesday in a company Q&A that it would release GPT-5.6 in limited preview form - granting access only to a small group of […]

TL;DR — When the spec is good, implementation needs less model. I started using a top-tier model to write the spec and a cheaper, faster one to implement it — still using the strong model, just spending it on the spec instead of the implementation. The gain isn't some magic prompt phrasing; it's the context: explicit business rules, audited project constraints, a defined output contract. That's systems engineering — the discipline of anyone who's kept real software alive, whatever their stack. Every backend dev knows the scene: the Swagger is out of date, the last hotfix shipped without a unit test, and the README.md documents a command nobody's used in six months. The code works. The docs lie. And the gap between the two is exactly where AI — and we — start to go wrong. I've spent the last few months developing with AI for real inside production projects, not tutorial greenfield. My takeaway was less about which model to use and more about a shift that already has a name: the move from prompt engineering to context engineering . The difference isn't semantic. Prompt engineering treats the problem as writing — finding the magic phrase. Context engineering treats it as what it always was: a systems engineering problem . And it's where my backend background applied most directly — though anyone who's kept a real system alive has the same instinct. The experiment that convinced me Let me start with the evidence, because that's what made me take this seriously. My reflex, for a long time, was to reach for the strongest model for everything — more expensive, smarter, fewer errors. Makes sense on paper. In practice, I saw something else. When the task's specification is well done — explicit business rules, audited project constraints, a defined output format — the model capability needed for implementation drops sharply. Enough to split the work by stage: I started using a top-tier model (currently Opus) to write the spec , and a cheaper, faster model (Sonnet) to implemen

The whole industry seems to have agreed on a standard for design tokens. The shift it sets up is still on its way. Design tokens are not new. The term was coined in 2014, at Salesforce, by Jina Anne and Jon Levine. 1 By 2017, Amazon had open-sourced Style Dictionary and the idea had spread well past Salesforce. We have been shipping design tokens for over a decade. What we never did, in all that time, was agree on a format. Every tool and every team rolled its own shape. There was never one neutral way to write a token down, its value and its meaning, so that any other tool could read it. Have you heard of DTCG? I hadn't, until recently. It is the Design Tokens Community Group, a W3C effort to finally settle that format. 2 The repo is quiet, but that is because the spec reached its first stable version in late 2025, not because anyone walked away. The quiet is a thing being finished, not abandoned. The list of who is backing it is not quiet at all. Adobe. Google. Microsoft. Meta. Amazon. Shopify. Salesforce. Sony. Pinterest. The New York Times. Disney. Framer. Penpot. Figma. Plus a dozen more. 2 That is not a side project. That is most of the industry quietly agreeing on something. One of those names, Figma , is the reason for the title of this piece. We will get to it, because the irony is the whole point. Here is my bet, and I will say up front that it is a bet. I think a storm is coming for design tooling. You do not have to believe me about the storm, because the bet does not depend on it. If you are wiring your tokens straight into one vendor's format, you are exposed. Anchor them to the open standard instead and you are not. The downside is lopsided. If I am wrong, you have lost almost nothing. If I am even half right, everyone hard-coded to a single tool is facing a rewrite. The format is young and already fragmenting. That is the point. The obvious objection is that the standard is too new to bet on, and already splintering. It is splintering. Google's DESIG

"Model unit prices keep falling, yet our monthly AI bill keeps climbing." If you use AI personally, you can feel the creep of your subscription and metered charges. If you own AI usage inside a company, the gap is even more pronounced. Overseas, this feeling has started getting a name: Tokenomics . On June 3, 2026, the Linux Foundation announced its intent to launch the Tokenomics Foundation , dedicated to open standards for AI cost management. Google, Microsoft, Oracle, JPMorganChase, and others — both providers and large buyers — are on board. https://www.linuxfoundation.org/press/linux-foundation-announces-the-intent-to-launch-the-tokenomics-foundation-to-establish-open-standards-for-ai-cost-management This post isn't an explainer of the word itself. It's an account of what changes for the people who own internal generative AI usage — the platform owners, the FinOps practitioners, the engineering leaders watching the bills — once you have this word in your vocabulary. What Tokenomics gives you isn't another saving technique. It changes the unit of measurement and the lens through which you read AI cost. Why Tokenomics, why now Tokenomics sits in the lineage of cloud FinOps. The FinOps Foundation now classifies Tokenomics as the "AI Value" dimension within FinOps for AI . Where cloud FinOps tracked the variable infrastructure costs (compute, storage, networking) against value, Tokenomics tracks the variable cost of intelligence itself. It's not a replacement; it adds a probabilistic, non-deterministic layer of variable cost on top. Tokens here means what you see on every API price sheet and usage dashboard — the smallest unit a language model reads and writes, the unit of compute. The word "tokenomics" also exists in the crypto world, but that one is about issuance, distribution, and incentives on a blockchain — tokens as units of ownership. Same word, different economies. https://www.finops.org/insights/token-economics-the-atomic-unit-of-ai-value/ The term gained

Autonomous AI agents are transitioning from experimental developer playgrounds into the core of enterprise application architecture. For organizations looking to automate complex workflows that require decision-making, reasoning, and tool use, agentic AI represents a paradigm shift. However, moving from a simple demo script to a reliable, production-ready enterprise agent system requires addressing significant architectural challenges. In this article, we will examine the core components of enterprise agent systems, design patterns for robust execution, and security considerations. The Core Architecture of an AI Agent An enterprise AI agent is more than just a large language model (LLM) loop. It is a system composed of four critical pillars: Reasoning & Planning (The Core LLM): The orchestrator that decides how to approach a problem, breaks down tasks, and analyzes output. Memory: Storing short-term execution traces (context) and long-term knowledge (vector databases, semantic memory). Tools (Action Space): APIS, databases, calculators, and code execution sandboxes that the agent can invoke to retrieve information or perform tasks. Guardrails & Evaluators: Decoupled verification layers that inspect the agent's plans and tool execution to enforce policy and security. +-------------------------------------------------------------+ | USER REQUEST | +-------------------------------------------------------------+ | v +-------------------------------------------------------------+ | AGENT ORCHESTRATOR / LLM LOOP | | * Planning (ReAct, Plan-and-Solve) | | * Memory retrieval | +-------------------------------------------------------------+ | ^ v (Call Tool) | (Tool Results) +------------------------+ +----------------------+ | TOOL ROUTER | | GUARDRAILS LAYER | | * APIs * Code Exec | | * Safety filter | | * DBs * RAG Lookup | | * Data sanitization | +------------------------+ +----------------------+ Planning Patterns: ReAct vs. Plan-and-Solve When designing how an agent re

Agent-testing startup Patronus AI, founded by former Meta AI researchers, is experiencing nearly insatiable demand, its investor says.

This week, Instagram launched a series of new features for its smart TV app that are all designed to get people to spend more time on the platform through the biggest screens in their homes. In addition to vertical Reels, Instagram for TV - which is currently available for Amazon Fire TV, Google TV, and […]

So long, Notion Mail.

The company said it is discontinuing its email inbox in favor of its AI agent offering as users are increasingly handing over the reins of their email to the agents.

Notion is "going all in on using agents to run your inbox."

Table of Contents Overview Southwest and AWS From On-Prem to Cloud AI Comes Into Play Kiro Why This Matters? Closing Thoughts Overview Southwest Airlines is one of the largest carriers in the world. Other than having by far my absolute favorite airplane livery, it's a massive enterprise based in Dallas, Texas, with more than 72,000 employees, over 4,000 daily flights during peak travel periods, roughly 134 million customers annually, and service across 120+ airports in 12 countries. At this scale, what really matters is operational reliability and speed. Because in aviation, slow operations cause delays, delays cause unhappy customers, and unhappy customers aren't particularly great for the company's revenue. A few minutes of latency in one system can turn into hours of disruption in the real world. So... can cloud and AI solve it? Absolutely, if done right. Southwest and AWS Southwest has selected Amazon Web Services (AWS) as its primary cloud partner to help modernize its technology stack and transform how the airline runs, develops systems, and serves its customers. Through this collaboration, Southwest plans to move away from a predominantly on-premises infrastructure toward a cloud-based, AI and agent-enabled architecture on AWS by 2028 . 2028 is not far away from now (Jun 25, 2026). This is very ambitious considering the amount of work that needs to be done. From On-Prem to Cloud Moving an enterprise of this size from on-prem infrastructure to the cloud is way more complex than it sounds. It doesn't sound easy either. It's one of the hardest things you can ever do as a software engineer, DevOps engineer, architect, engineering manager, or anyone else involved in it. It involves a lot of steps, including but definitely not limited to: Assessing existing systems, dependencies, and infrastructure to understand what needs to move and how Defining a migration strategy (lift-and-shift, replatforming, or full refactoring to cloud-native architecture) Designing and bu

The problem Just saying "hi" to Claude Code costs ~31,000 tokens . I was paying $500+/month in API costs and had no idea where the tokens were going. So I built tokenwise — a free CLI that shows exactly where your AI coding agent wastes tokens. What it does tokenwise audit — Scan your instruction files It scans your CLAUDE.md, AGENTS.md, and rules files, then shows: How many tokens each file costs per message Boilerplate the AI already knows ("Always write clean code") ALL-CAPS emphasis that doesn't help (NEVER, ALWAYS, MUST) Duplicate sections Unscoped rules that load when they shouldn't tokenwise scan — Analyze your sessions It reads your latest session logs and shows: Token breakdown by component (system prompt, history, tool output) Cache hit rate Top 3 "token hogs" with actionable tips Monthly cost projection The key insight Most people try to compress context (which makes the AI dumber). tokenwise measures first, then applies safe fixes. You can't optimize what you can't measure. Quick start npx @davizin713/tokenwise audit npx @davizin713/tokenwise scan Zero API calls. Zero LLM inference. 100% local. Free forever. Works with 11 agents Claude Code, OpenCode, Cursor, Aider, Cline, Codex CLI, Goose, Continue.dev, Windsurf, Augment, and Kilocode. Links GitHub: github.com/davi713albano-coder/tokenwise npm: npmjs.com/package/@davizin713/tokenwise Built with TypeScript / Node.js js-tiktoken for token counting sql.js for reading OpenCode sessions MIT licensed If you use Claude Code or any AI coding agent, try it and let me know what you think! ⭐

It took 20 years, but the Finance app arrives just in time to be packed full of AI.

Provisioning a Bedrock RAG knowledge base with S3 Vectors, without the hallucinated API calls. If you've asked an AI coding agent to set up AWS, you've seen it confidently invent a parameter, reach for a deprecated service, or burn ten minutes retrying against a service it never saw in training. The failure mode that bites hardest is the silent one: the agent thinks it succeeded, and you find out an hour later. I hit two of these while standing up the retrieval layer for a LangGraph support bot, an Amazon Bedrock Knowledge Base backed by Amazon S3 Vectors. I'd love to say I caught both with deep AWS expertise. I caught them because the Agent Toolkit for AWS read the docs I hadn't. Both would have shipped, and neither did. The 30-second setup The goal: take a folder of markdown product docs and make them queryable by meaning, so an agent can answer "is this safe for color-treated hair?" from the real docs instead of guessing. Think of it as giving the agent a library it can search instead of making things up. That's the retrieval half of RAG, the foundation a LangGraph agent will later call as a tool. Four moving parts, wrapped in one managed service: Source bucket : an S3 bucket holding the docs. Embeddings : Amazon Titan Text Embeddings V2 (1024-dim vectors). Vector store : Amazon S3 Vectors. I chose it over OpenSearch Serverless because it has no always-on compute, the difference between cents and a monthly surprise for a demo that sits idle. Knowledge Base : Amazon Bedrock Knowledge Bases ties it together into one thing you can query with a retrieve call. To follow along, you need an AWS account, a non-root IAM identity with credentials configured locally, uv installed, and the toolkit installed in your agent. The fastest path across Kiro, Claude Code, Cursor, and Codex is the AWS CLI installer, aws configure agent-toolkit ; in Kiro you can instead add the AWS MCP Server to .kiro/settings/mcp.json (pin the mcp-proxy-for-aws version) and run npx skills add aws/age

Disclosur: Ini dari tim Nexotao — saya bahas gateway kami sendiri di bawah. Saya jaga sebatas fakta yang bisa kamu cek sendiri: semua nama model, context window, dan harga ada di halaman pricing kami, dan saya kasih linknya. Kalau kamu developer di Indonesia, kemungkinan besar pernah kejedot ini: API OpenAI dan Anthropic minta kartu kredit luar negeri . Nggak punya kartu, nggak bisa pakai API. Banyak dari kita mentok di situ. Solusi yang jalan sekarang: gateway lokal yang nerima QRIS / Rupiah . Ini versi jujurnya — gimana cara kerjanya, berapa biayanya, dan apa yang belum bisa. Dua model live, satu API yang kompatibel Lewat Nexotao kamu pakai dua model teks: Claude Opus 4.8 ( claude-opus-4-8 ) — context window 350.000 token DeepSeek-V4-Pro — context window 131.072 token Itu angka context window yang dipublikasikan apa adanya — tanpa pemotongan diam-diam. Endpoint-nya kompatibel dengan OpenAI dan Anthropic , jadi biasanya cukup ganti base URL sama key-nya. Format OpenAI: from openai import OpenAI client = OpenAI ( base_url = " https://api.nexotao.com/v1 " , api_key = " sk-nexo-... " ) resp = client . chat . completions . create ( model = " claude-opus-4-8 " , messages = [{ " role " : " user " , " content " : " Halo " }], ) print ( resp . choices [ 0 ]. message . content ) Format Anthropic: curl https://api.nexotao.com/v1/messages \ -H "x-api-key: sk-nexo-..." \ -H "anthropic-version: 2023-06-01" \ -H "Content-Type: application/json" \ -d '{"model":"claude-opus-4-8","max_tokens":256, "messages":[{"role":"user","content":"Halo"}]}' Cara bayarnya Top up saldo Rupiah via QRIS , mulai Rp10.000 . Tanpa kartu luar negeri. Bayar sesuai pakai — dipotong per token. Tanpa langganan , dan saldo nggak hangus. Tiap response ada header X-Cost-Rp , jadi kamu lihat biaya rupiah persis tiap request. Berapa biayanya Saat tulisan ini dibuat, Claude Opus 4.8 lewat gateway sekitar 70% lebih murah dari harga retail resmi (input) — tapi jangan percaya saya gitu aja. Halaman perbandingan har

Disclosure: This is the Nexotao team — I'm describing our own gateway below. I've kept it to facts you can verify yourself: every model name, context window, and price here is on our live pricing page, and I link it. If you're an Indonesian developer, you've probably hit this wall: the OpenAI and Anthropic APIs want a foreign credit card . No card, no API. A lot of us get stuck right there. The fix that works today: a local gateway that takes QRIS / Rupiah . Here's the honest version of how it works, what it costs, and what it doesn't do. Two live models, one compatible API Through Nexotao you call two text models: Claude Opus 4.8 ( claude-opus-4-8 ) — context window 350,000 tokens DeepSeek-V4-Pro — context window 131,072 tokens Those are the real, published context windows — no silent truncation. The endpoint is OpenAI- and Anthropic-compatible , so you usually just change the base URL and key. OpenAI format: from openai import OpenAI client = OpenAI ( base_url = " https://api.nexotao.com/v1 " , api_key = " sk-nexo-... " ) resp = client . chat . completions . create ( model = " claude-opus-4-8 " , messages = [{ " role " : " user " , " content " : " Hello " }], ) print ( resp . choices [ 0 ]. message . content ) Anthropic format: curl https://api.nexotao.com/v1/messages \ -H "x-api-key: sk-nexo-..." \ -H "anthropic-version: 2023-06-01" \ -H "Content-Type: application/json" \ -d '{"model":"claude-opus-4-8","max_tokens":256, "messages":[{"role":"user","content":"Hello"}]}' How you pay Top up your Rupiah balance via QRIS , from Rp10,000 . No foreign card. Pay-as-you-go — deducted per token. No subscription , and the balance never expires. Every response carries an X-Cost-Rp header, so you see the exact rupiah cost of each request. What it costs At the time of writing, Claude Opus 4.8 runs roughly 70% below official retail input pricing through the gateway — but don't take my word for it. The comparison page shows live per-token rates and computes "vs official" automati