标签：#ai

Nvidia said Hugging Face hosts over 3 million models and is used by over 18 million developers.

TL;DR — On a MSI Stealth A16 AI+ (Ryzen AI 9 365, XDNA2 NPU) running Arch, I got OpenAI's whisper-large-v3-turbo transcribing on the NPU — not the CPU, not the GPU — at RTF ≈ 0.18 (a 30 s clip in ~5.2 s) for roughly a tenth of the energy the same job costs on the CPU, plus an LLM answering on the same NPU through an OpenAI-compatible API. The whole path is local and offline. This is the write-up of the driver stack, the one real gotcha (memlock), and the runtime that made it a 20-minute job instead of a weekend. Why this is worth writing down AMD's "Ryzen AI" NPU (the XDNA / XDNA2 block in Phoenix / Hawk Point / Strix Point laptops) is marketed almost entirely around Windows: the Ryzen AI SDK, the ONNX Runtime VitisAI execution provider, Lemonade, and the demos all assume you're on Windows with the official stack. On Linux the picture in early 2026 is better than most people think — the NPU driver has been in the mainline kernel as amdxdna since 6.14 — but the "load a real model and run it" story still isn't well documented. Here's what actually worked, end to end. The hardware Part Detail Laptop MSI Stealth A16 AI+ A3HVGG APU AMD Ryzen AI 9 365 (Strix Point) NPU XDNA2, 8 columns, exposed as /dev/accel/accel0 NPU firmware 1.1.2.64 Kernel 7.1.9-arch1 ( amdxdna in-tree) OS Omarchy (Arch Linux) AMD quotes the Strix Point NPU at up to 50 TOPS, INT8 . 1. The driver stack Three pieces have to be in place before any runtime can touch the NPU: amdxdna — the kernel driver. In-tree from Linux 6.14; it's what creates /dev/accel/accel0 . Check it's bound: $ ls /dev/accel/ accel0 $ dmesg | grep -i amdxdna XRT (Xilinx/AMD Runtime) + the xrt-plugin-amdxdna shim. XRT is the userspace API; the plugin teaches it about the XDNA device. On Arch both are in extra : $ sudo pacman -S xrt xrt-plugin-amdxdna $ xrt-smi examine ... XRT Version : 2.21.75 NPU Firmware Version : 1.1.2.64 Device(s) Present |BDF |Name | |----------------|--------------| |[0000:66:00.1] |RyzenAI-npu4 | You want a D

There is a strange contradiction happening in software engineering right now. A lot of developers are worried that AI is going to make them obsolete. At the same time, the people building the most capable AI coding tools are demonstrating something that should probably make us rethink what being a software engineer actually means. I don't think the future is one where nobody understands software anymore. I think it is one where writing the software becomes dramatically cheaper. And if that happens, the thing that makes an engineer valuable has to move. That is what I mean by career safety. Career safety isn't about making yourself impossible to replace. It is about making your value portable. We've always resisted giving up the code Developers have a long history of being suspicious of abstractions that take work away from us. We went from machine code to assembly, from assembly to higher-level languages, from manually managing memory to garbage collection, from building everything ourselves to libraries and frameworks, and from text editors to IDEs. We even had entire categories of tools, such as CASE tools, designed to automate parts of software development. And every time, there was resistance. Because programmers don't just use code. We build our identities around it. John Carmack captured this unusually well when he wrote: “Coding” was never the source of value, and people shouldn’t get overly attached to it. — John Carmack He followed that with the more important point: Problem solving is the core skill. — John Carmack That is a difficult idea for developers to internalize because coding is tangible. You can point at the repository. You can point at the pull request. You can count the commits. You can say, "I wrote this." But the business doesn't ultimately pay you for the number of lines you wrote. It pays you for what those lines accomplish. The business never really bought the code A company doesn't wake up in the morning thinking: "We need 14,000 more line

NVIDIA has confirmed it is buying AI software repository Hugging Face.

Zoox is expanding to this critical ride-hailing destination a few weeks after it started charging for robotaxi rides.

Nvidia has agreed to buy Hugging Face for $12.93 billion, bringing one of the most popular hosting platforms for open-source AI models, datasets, and tools under the ownership of the world's biggest AI chipmaker. Hugging Face is an online platform founded in 2016 that gives AI developers a space to share their projects and data […]

Restaurant menus with AI-generated food images are going viral for their off-putting aesthetic. But AI’s impact on modern menus runs even deeper.

These WIRED-tested indoor and outdoor air quality monitors give you insights into your home’s air quality, warn of potentially harmful pollutant levels, and help keep tabs on your environment.

Gemini reportedly told a group of climbers that it would only take them eight hours to climb Mt. Shasta. They had to be rescued more than a day later.

Most "AI coding" products still put a chat window next to your editor and call it a day. I wanted something closer to what Claude Code already does well on a server: give it a real computer and let it drive. Superagent is a Mac app that gives Claude Code (or any agent you point it at) an actual environment to work in, not just a text box. Concretely: A real browser it can navigate, click, type into, and read the DOM of, not screenshots and guesses. An iOS Simulator window it can install apps into, tap through, and screenshot to verify UI changes. A relay that pairs your Mac with your phone, so the agent can keep working (and you can keep watching) from your pocket. The core idea is boring on purpose: don't build a smarter chat window, build a better place for the agent to act. Most of the interesting failures I hit while building this weren't in the model, they were in the environment: synthetic file inputs that don't persist through a web form's upload component, elements that exist in the DOM but aren't in the accessibility tree, simulator state that drifts from what a screenshot shows. Fixing those is what actually makes an agent reliable to hand a task to. It's built as three pieces: an Electron desktop app, a SwiftUI iOS companion, and a small Cloudflare Worker relay that pairs the two with per-address rate limits so a lost phone can't be used to spam a stranger's Mac. If you want to see it: https://peerlist.io/pungme/project/superagent-for-mac Happy to answer questions about the browser automation approach, the simulator driving, or the relay's pairing/security model in the comments.

An AI coding agent working inside Claude Desktop or Cursor can read your code, write new files, and run your test suite — but it can't open a browser, log into a dashboard, and click "generate" to get a batch of realistic test data. It has no hands for a UI. AI agent test data generation only works if there's something the agent can call : a tool with a defined schema it can invoke mid-session, the same way it calls a file-write or a shell command. That's exactly what the Model Context Protocol (MCP) is for, and it's why we shipped @jsonfabrica/mcp-server on npm. What AI agent test data generation requires over MCP MCP lets an AI client — Claude Desktop, Cursor, or anything else that speaks the protocol — launch a small local server over stdio and treat its exposed functions as tools it can call during a conversation. The agent decides when to call jsonfabrica_generate_from_template the same way it decides when to call read_file . For that to work, three things have to exist: a server process the client can start, a set of tool definitions with typed inputs and outputs, and — underneath all of it — some actual operation the tool call triggers. MCP server test data generation is that last piece: the tool call has to result in real, schema-conformant data coming back, not a stub. @jsonfabrica/mcp-server , concretely We published @jsonfabrica/mcp-server v0.1.1 as a local MCP server: the AI client launches it itself over stdio, no separate process to manage, no port to open. It exposes the JsonFabrica gateway as a set of MCP tools — jsonfabrica_create_template , jsonfabrica_generate_from_template , jsonfabrica_generate_adhoc , jsonfabrica_create_batch , jsonfabrica_create_sequence , and more. Mid-session, an agent can create a template matching the shape of your User or Order model, generate a batch of realistic records against it, and drop the result straight into a fixture file or a seed script — without you leaving the editor to go configure anything by hand. Why thi

Anker has a bunch of new AI-infused headphones, earbuds and hearing aids on the way.

An agent patch can pass every unit test and still write outside the workspace, call an undeclared tool, or read an env key the task never named. Gate the blast radius first. Score the prose later. This article is a method, not a field report. It proposes a fail-closed envelope around filesystem roots, tool names, environment keys, and network hosts. Side-effect violations never freeze. Only a dual-runner disagreement on a non-envelope property may freeze, and only with a hashed evidence bundle. The conclusion in one rule Treat an agent patch as a capability change. If the run touches anything outside a declared envelope, the gate fails closed. Flakes in ranking, wording, or latency do not override that rule. Cheap generation does not make side effects cheap to reverse. A green suite that never watched /tmp , os.environ , or outbound sockets is not a verification result. It is a missing observer. What this gate is not It is not a golden-file of model text. It is not a mutation score. It is not a full-suite rerun after every hunk. It answers four questions only: Did the run write or delete outside allowed roots? Did it invoke a tool name that is not on the allowlist? Did it read an environment key that is not on the allowlist? Did it open a network host that is not on the allowlist? If any answer is yes, fail. Do not freeze. Do not retry for luck. Artifact: a locked envelope and an observer log Pin the envelope as a fixture. Hash it. Refuse to run if the hash drifts without a review note. { "envelope_id" : "agent-patch-envelope-v3" , "allowed_roots" : [ "/work/repo" , "/tmp/agent-scratch" ], "allowed_tools" : [ "read_file" , "apply_patch" , "run_tests" ], "allowed_env" : [ "CI" , "RUN_ID" , "ENVELOPE_HASH" ], "allowed_hosts" : [], "network" : "deny" } sha256sum envelope.json > envelope.json.sha256 # CI must compare this digest before the agent process starts. Label the next block as a proposed harness, not a production sandbox. User-space tracing will miss kernel-leve

Untrusted retrieval is now a more common production failure than a weak prompt, because agents ingest memory they never score. A seventy-minute workshop can add a cheap scoring gate, a replayable log, and a reject path before generation. Students leave with a runnable Python harness, a four-row decision table, and a timing plan they can repeat. The method stays useful if every product name is removed and the scoring host is only a free server. What you will build This workshop treats retrieved snippets as untrusted input, not as ground truth the model should quote. You will capture a retrieval batch, score each chunk against a written rubric, and allow only passing chunks into the prompt. A JSONL replay log records the fingerprint, score, and decision so later failures can be diffed. The generation model never sees dropped text, which keeps stale or planted memory out of the answer. Timing box 00:00–00:10 — install dependencies, copy the harness, and load the sample corpus 00:10–00:30 — Exercise 1: capture retrieval payloads and stable fingerprints 00:30–00:50 — Exercise 2: score chunks with a rubric and an optional free model 00:50–00:65 — Exercise 3: gate the prompt and replay one rejected case 00:65–00:70 — debrief against the decision table and list remaining holes The schedule is a teaching box, not a production SLA, and it assumes one laptop plus one HTTP scoring endpoint. If the endpoint is slow, freeze Exercise 2 after five scored chunks and continue with the logged samples. Do not expand the window to chase a perfect judge; the learning goal is a gate you can rerun. Why a scoring pass belongs in front of generation Cheap code generation has made it easy to wire a retriever into a chat loop in an afternoon. The failure mode that follows is quieter than a crash: the model answers fluently from a chunk that is expired, off-topic, or injected. Architecture diagrams rarely show that hop as a trust boundary, so teams skip scoring and jump to a larger generator. A

Master Prompts in 2026: Stop Prompting Like It's 2023 I still see people paste a 40-line “act as a senior expert with 20 years of experience” block into ChatGPT and call it engineering. That stopped working as a strategy a while ago. Models got better. Context windows got bigger. Agents started calling tools. And the failure mode shifted. It’s rarely “the model is dumb” now. It’s “your system has no contract.” This is a long, practical write-up on master prompts — the stable policy layer above individual tasks. How to write them. How to force planning. How to run Plan → Act → Observe → Verify without theater. How to make the same prompt useful to a tired human at 11pm and to an agent loop that only understands schemas. I’ve broken enough production prompts across GPT-4o, Claude 3.5 Sonnet, and Gemini-class stacks to have opinions. Some of them are uncomfortable. TL;DR / Key Takeaways A master prompt is not a clever sentence. It’s the policy layer : role, success criteria, process, constraints, output contract, failure handling. Production reliability comes from LLM orchestration patterns — plan JSON, single-task executors, and explicit done_when checks — not from longer personality blocks. JSON contracts + verification beat free-form answers. Agents that can’t prove completion will invent it. Treat prompts like code: version them, eval them, and put a real verify step after generation (including SEO/quality checks when you publish). Table of Contents What a master prompt actually is The 7-part anatomy that doesn’t collapse under pressure Frameworks worth keeping (and which ones to ignore) Planning is the real skill From plan to agent loop Context engineering beats clever wording Few-shot, JSON contracts, and the anti-hallucination rule Copy-paste masters you can actually deploy A real publish pipeline (including the verify step people skip) Eval or you’re guessing Failure patterns I keep seeing PromptOps: treat prompts like code One universal master prompt Ship chec

Do you ever feel like the AI world has moved forward a little too quickly? You hear about self-healing systems and autonomous agents and start wondering whether we've already built Skynet or everyone around you is just messing with you. When someone mentions HITL or MCP, you no longer know whether it's some secret code used by an AI cult or maybe the stage names of famous DJs. You're not alone! 😉 In this article, I'm deliberately using a lot of simplifications. My assumption is simple: either you already know these terms and don't need another five-paragraph academic explanation, or you don't really know what they mean. And in that case, the last thing you need is an academic definition. And yes, this is already my third listicle in a row. Believe me, this is NOT some growth hacking strategy xDDD. Pure coincidence. It just so happens that in two weeks (HOLY SH*T!!!), I'll be speaking at AGNTCon + MCPCon Europe , where I was invited because of this wonderful article that I wrote here on DEV. I swear I had at least as much fun writing it as people apparently had reading it. So yes, I know WebMCP reasonably well, but I'm planning to attend a lot of other talks there too, so apparently a refresher won't hurt me either. 😅 Anyway, back to the point. I strongly believe that people remember things best through examples. And for many people, the ultimate examples are rich people, otherwise known as successful people . So let's imagine that our hypothetical protagonist is very, very, disgustingly rich. He's actually a billionaire. He earned his fortune through hard work and by running several companies. He makes cars and rockets, bought his own social media platform, and recently even acquired an AI coding company. A person like this would obviously need his own AI agent. And because our protagonist needs a name, let's call him Elon Mózg . Mózg means brain in Polish, which works beautifully here. One more thing: my examples could probably also serve as prompts for a coding ag

Cohere has launched Parse 5, a multimodal foundation model designed to extract structured data from complex enterprise documents. The 2.3-billion-parameter system converts visually rich PDFs into Markdown while providing bounding box coordinates for visual grounding. It has been evaluated against over 2,000 enterprise pages, achieving an average score of 79.2 in key performance areas. By Olimpiu Pop

Imagine an HR manager who has just completed this month’s hiring drive. Dozens of candidates have accepted their offers, and everyone is eager to onboard them as quickly as possible. But before that can happen, every background verification request has to move through multiple checks, vendors, approvals, and status updates. What should be a straightforward operational process often turns into hours of chasing spreadsheets, checking vendor portals, and answering the same status questions over and over again. It isn’t a technology problem as much as it is a visibility problem — and that’s the challenge that inspired BGVerifyIQ. The problem every HR team quietly struggles with Ask any HR or Talent Acquisition team what happens after an offer letter goes out, and you’ll hear a familiar story. A candidate is selected. A background verification (BGV) request is raised with a vendor. And then… the waiting begins. Behind that wait is a surprising amount of manual work: HR has to log into multiple vendor portals just to check where a case stands. Status updates live in scattered Excel sheets, emails, and vendor dashboards that rarely agree with each other. Nobody has a single, live number for “how many candidates are actually ready to be approved today.” SLA breaches — cases that have been open too long — get discovered late, usually when a hiring manager complains. When a discrepancy or a “suspect” flag shows up, it can get buried in a spreadsheet row instead of being escalated immediately. Every time leadership asks “how are our BGV vendors performing,” someone spends half a day pulling numbers together manually. New joiners get onboarded before their verification is even reviewed properly, simply because nobody had a clear, current view of their case. None of this is because HR teams aren’t diligent. It’s because the tools weren’t built for this specific job. Spreadsheets don’t talk to vendor systems. Vendor portals don’t talk to each other. And nobody has time to become

I sat in on Margaret-Anne Storey's DORA community session last week, and she put a name on the thing I'd been circling since April. It isn't technical debt. Her ACM Queue piece splits software health into three debts. Technical debt is the familiar one: implementation choices that make tomorrow's change harder. Intent debt is the missing rationale, the goals and constraints that say what a system is even for, which now has to be legible to agents and not just to people. Cognitive debt is the one that stopped me. It's the erosion of shared understanding, the state where nobody on the team can confidently explain how the system works or predict what a change will break. Read that again if you review pull requests for a living. I closed a thirteen-post retrospective last month admitting I couldn't answer one question: how many AI-generated pull requests a week can a review process absorb before it stops working as a control? I still don't have that number. What I have now is a name for what you accumulate while you don't have it, and a way to find yours. Approval velocity measures motion Every metric most teams watch gets better as review collapses. Merge rate climbs. Time-to-approve drops. The throughput chart looks terrific right up until the incident review, because a reviewer who has quietly become a rubber stamp is indistinguishable from a fast reviewer in every dashboard you own today. Cognitive debt doesn't announce itself as a red number. It shows up as green ones, arriving faster. I know this failure mode from the inside. Two months of green CI on conformance checks that had never once passed , on my own project. A human audit caught it. No metric I was watching came close. What you need to measure is detection. Almost nobody does. Mutation testing, pointed at the reviewers We solved this once already, for test suites. Mutation testing injects known bugs into code and checks whether the tests catch them. A suite that passes everything might be thorough or migh

AI apps have moved past simple chat boxes. Today's AI systems need agents, tools, memory, data, security, monitoring, and scale. The hard part is not calling an LLM API. The hard part is building a reliable system around that API call. 1. From LLM Demo to Production System A demo is simple: flowchart LR A[Prompt] --> B[Model] --> C[Response] A real production system looks very different: flowchart TD U[User] --> API[API] API --> APP[Application Layer] APP --> ORCH[AI Orchestration] ORCH --> LLM[LLM] ORCH --> TOOLS[Tools] ORCH --> RAG[RAG] ORCH --> MEM[Memory] ORCH --> GUARD[Guardrails] ORCH --> DATA[Data + Infrastructure] DATA --> OBS[Observability] Each box matters. If you skip Guardrails , bad input can hijack your system. If you skip Memory , every message re-explains itself and costs more tokens. If you skip Observability , you won't know why the system failed until a user tells you. The rest of this article walks through each box. 2. Where AWS Fits Instead of listing AWS services, let's match each one to a real problem. Problem AWS Service Why Need a foundation model Amazon Bedrock Managed access to multiple LLMs, no infra to run Store documents and files S3 Cheap, durable, scales easily Store app data RDS / Aurora / DynamoDB Structured data, users, sessions, transactions Search by meaning (retrieval) OpenSearch / pgvector Vector search for RAG Run code Lambda / ECS Serverless or container compute for your app logic Handle async work SQS / EventBridge Queue jobs, decouple slow tasks, avoid lost requests Watch the system CloudWatch Logs, metrics, alarms Keep it secure IAM / Secrets Manager Access control and safe storage of keys flowchart LR subgraph Compute L[Lambda / ECS] end subgraph Data S3[(S3)] DB[(RDS / DynamoDB)] VEC[(OpenSearch / pgvector)] end subgraph AI BR[Bedrock] end subgraph Ops CW[CloudWatch] SEC[IAM / Secrets Manager] end L --> BR L --> S3 L --> DB L --> VEC L --> CW L --> SEC 3. AI Agents Change the Architecture An agent doesn't just answer — i