AI is moving into the background—and putting the infrastructure beneath it in the spotlight

AI is moving into the background—and putting the infrastructure beneath it in the spotlight.

Agents need more than intelligence

September 15’s coverage highlights the operational demands of AI that keeps working after a user steps away. Pizza Bot offers an inbox for background tasks, separating completed work from runs awaiting decisions. A discussion with platform specialist Mauricio Salatino explores supporting agentic systems, while OpenShell examines formal methods for reasoning about permission changes. Together, these pieces frame a central challenge: making autonomous work manageable, durable, and accountable.

Performance becomes a full-stack pursuit

Hardware and inference efficiency dominate another cluster of stories. NVIDIA Vera validation reports 1.64× performance per core against a leading x86 processor on the tested agentic CPU workloads. Other articles explore DSpark training for Kimi-K3 on GB300 NVL72 hardware and running DeepSeek-V4-Flash with a million-token context across two DGX Sparks. Approaching.AI’s account of using Mooncake to serve over a trillion tokens daily adds a production-scale perspective.

Beyond throughput, knowledge graph integration with Poro2 explores enriching medical text processing, while NVIDIA FLARE coverage addresses the infrastructure challenges of scaling federated learning.

Developer velocity meets systems reality

An account of building a Linux GPU driver for the M4 Mac Mini in one month sits alongside experiments adapting Git packfiles for object storage and Lovable’s planned preview improvements. Dart and Flutter’s Google Summer of Code results add a community-driven development story.

Yet faster delivery still depends on understanding underlying constraints. Articles on Intel’s subnormal-number costs, iOS memory-pressure terminations, and serverless connection pools expose easily overlooked bottlenecks. The day’s connecting thread: ambitious software increasingly depends on careful work below the interface.

Featured Articles

Show HN: Pizza Bot – An inbox for AI agents that work in the background

Pizza Bot is an inbox for long-running AI work. Start or schedule a task, return to your day, and let completed work collect in Unread while runs waiting for your decision collect in Action. Agents ke...

Keywords: pizza bot, process pizza, bot guide, running pizza, bot configuration, bot uses, cron webhook, running api, bot developed, make pizza
Source: github.com

Validating NVIDIA Vera on agentic CPU workloads

Validating NVIDIA Vera on agentic CPU workloads: 1.64x per core against a leading x86 processor

Ryan Shrout Most of the time today, every discussion of AI infrastructure starts with the GPU, and for...
Keywords: agentic cpu, agentic workloads, agent throughput, vera cpu, cpu vera, core performance, cpu execution, performance agentic, cores nvidia, processor task
Source: signal65.com

Building a Linux GPU Driver for the M4 Mac Mini in One Month

I Came, I Prompted, I Left Part 2: Building a GPU Driver From Scratch in One Month Previous blog post: https://codyho.dev/blog/hypervisor-macbook-neo/ What We Did TL;DR: Niklas and I built a fully Ope...

Keywords: apple gpu, refactor gpu, building gpu, gpu effectively, driver gpu, gpu submissions, gpu driver, gpu submission, gpu drivers, writing gpu
Source: codyho.dev

Knowledge Graph Integration With Poro2 For Enriching Medical Text Processing

Knowledge Graph Integration With Poro2 For Enriching Medical Text Processing# Medical documentation contains specialized terminology, such as terms like “hepatomegalia” or “bilateraalinen pleuraeffuus...

Keywords: terminology services, terminology tools, terminology tool, knowledge language, terminology model, dictionary provides, dictionary knowledge, clinical language, curated terminology, term language
Source: rocm.blogs.amd.com

Comparing Open Source LLMs in 2026: Llama 3, Mistral, and Gemma

As businesses and developers increasingly rely on Large Language Models (LLMs) for tasks ranging from text generation to complex data analysis, choosing the right suite of tools becomes more crucial t...

Keywords: specialized nlp, nlp tasks, deep learning, language models, nlp applications, natural language, processing nlp, large language, knowledge llms, enhanced language
Source: collabnix.com

How we trained the fastest DSpark for Kimi-K3 using GB300 NVL72

How we trained the fastest DSpark for Kimi-K3 using GB300 NVL72 In June, DeepSeek shipped DSpark, an extension to the DFlash block-level speculative decoding algorithm. The new algorithm promised stro...

Keywords: fastest dspark, k3 dspark, dspark speculator, dspark training, dspark kimi, dspark draft, dspark architecture, inference dspark, new dspark, dspark
Source: vllm.ai

You can run git on object storage if you re-make packfiles

You can run git on object storage if you re-make packfiles Git packfiles were designed for mmap and local disk, so pulling one object out of a bucket means guessing at a byte range. I wrote a new form...

Keywords: storing git, git objects, filesystem git, git object, objects git, git uses, git packfiles, git large, packfiles git, pack git
Source: tigrisdata.com

Two DGX Sparks, One Model: DeepSeek-V4-Flash with a 1M-Token Context

I got some new hardware. Two NVIDIA DGX Sparks. I connected them to each other with two QSFP cables, and put both of them on my Tailscale network. A single Spark has 128 GB of memory, shared between t...

Keywords: dgx spark, dgx sparks, vllm sparks, spark vllm, spark gpu, runs deepseek, spark setup, vllm deepseek, spark 128, deepseek v4
Source: dashaun.com

Google Summer of Code 2026 results

Google Summer of Code 2026 results Today, we spotlight the projects and results from contributors mentored by Dart and Flutter in Google Summer of Code 2026. The Google Summer of Code (GSoC) program f...

Keywords: dart developer, dart google, flutter google, new developers, github dart, development google, dart project, flutter participated, developers worldwide, developers
Source: dart.dev

P99 CONF Speaker Spotlight: Daniel Lemire

Persistence apparently pays off. The P99 CONF community has been requesting Daniel Lemire for a few years now, and the stars finally aligned. In just about a month (!), Daniel and Francisco Geiman Thi...

Keywords: billion queries, queries, hashing, perfect hashing, compile time, fast hash, large codebases, lookups, talking p99, lookup
Source: p99conf.io

Scaling Federated Learning Across Docker, Kubernetes, and Slurm with NVIDIA FLARE

Federated learning (FL) projects often begin with a straightforward setup: one server, a few clients, and one dataset at each site. As those projects grow, the challenge shifts from running an algorit...

Keywords: federated learning, persistent federation, learning infrastructure, gpu clusters, infrastructure gpus, site federation, federation use, federation processes, maintain federation, federated workloads
Source: developer.nvidia.com

Building Platforms for AI Agents, with Mauricio (Salaboy) Salatino

Brian Teller: In this episode of Kube Signals, I spoke with Mauricio Salatino about what changes when platform teams move from running traditional cloud-native applications to supporting agentic syste...

Keywords: ai infrastructure, infrastructure agents, agentic platform, agent systems, understand agents, understanding agents, enables agents, agentic frameworks, agents trust, agentic applications
Source: ku.bz

Faster previews, soon powered by OJ

A Lovable preview is not a static page. It is a real developer server running your real app, with hot reloading. That means it keeps the app running and sends your browser new versions of the files yo...

Keywords: running previews, preview static, previews run, previews production, lovable previews, lovable preview, previews faster, preview usable, preview use, vite apps
Source: lovable.dev

Model Management

The previous article went through the Kronk SDK: a batching engine, a conversation cache, and a pile of workarounds behind a very small public API. All of it rested on three assumptions: that the mode...

Keywords: kronk models, models kronk, kronk model, memory kronk, kronk vram, load kronk, kronk sdk, sdk kronk, kronk defaults, file kronk
Source: internals-for-interns.com

Subnormal floating-point numbers are expensive… on Intel processors

15 September 2026 · 4 min We represent floating-point numbers using the IEEE standard. For very small numbers, the standard uses special subnormal numbers. Unfortunately, they have a reputation of mak...

Keywords: subnormal slow, numbers performance, slower multiplication, subnormal latency, processors subnormals, benchmark kernels, subnormal compiler, computing subnormal, subnormal values, values subnormal
Source: lemire.me

Why iOS Killed Your App Without Crashing: Jetsam, Virtual Memory, and Darwin Memory Pressure

Why iOS Killed Your App Without Crashing: Jetsam, Virtual Memory, and Darwin Memory Pressure Your app disappeared and there is no crash report. We trace memory from a Swift allocation down through Dar...

Keywords: xcode crashes, app crashing, xcode memory, memory swift, ios memory, investigate crash, crashes crash, ios killed, crashing, crashes
Source: kyawthemonkey.com

Authentication is a long-running Workflow

In this article#

A login is not a request
Model the connection, not the attempt
The human is part of the Workflow
Staying logged in is part of the product
Go owns durability; TypeScript runs...
Keywords: authentication maintained, automatic reauthentication, managed auth, authenticated session, auth agent, auth remember, authentication long, auth model, logged authentication, authentication
Source: temporal.io

celld: Durable Objects, Unbundled

-5b301f10ddcd---4 crawled_date: 2026-09-15T19:34:58.582326+00:00 feed_url: https://itnext.io/feed published: Tue, 15 Sep 2026 19:00:32 GMT

<figure><img alt="" src="https://cdn-images-1.medium.com...

Keywords: entities agents, agent architecture, ai agent, ai agents, entity agent, autonomous entities, example agent, containing agent, model agent, stateful entities
Source: itnext.io

Over One Trillion Tokens per Day: How Mooncake Powers Approaching.AI’s Token Factory

Over One Trillion Tokens per Day: How Mooncake Powers Approaching.AI’s Token Factory This article was contributed by Approaching.AI, sharing how it uses Mooncake to build large-scale LLM inference inf...

Keywords: ai tokens, ai token, token factories, token production, trillion tokens, token factory, trillion token, tokens wasting, token demand, ai mooncake
Source: kvcache.ai

Serverless Killed Your Connection Pool

Serverless Killed Your Connection Pool A connection pool is one of those things you set up once and stop thinking about. Ten connections, reused forever, and the cost of opening one disappears into th...

Keywords: connection pooling, connection pool, pooled connection, postgres pooler, average serverless, pool connection, pool connectionstring, pooler latency, pooler cost, query serverless
Source: devops-daily.com

What we have learned at OpenShell applying formal methods to control AI agents

What we have learned applying formal methods to control AI agents An intro to using formal methods to reason about permission changes in long-running AI agents. In this post- we’ll dive into how permi...

Keywords: agents need, ai agents, agent control, ai agent, agents smarter, agent needs, agent policy, agent permissions, supervision agents, agent write
Source: nvidia.github.io

I Tried 20+ AWS SysOps Administrator Associate Certification (SOA-C03) Courses, Practice Tests, Mock Exams, and Dumps - Here Are My Top 5 Recommendations

Hello guys, if you are preparing for AWS certifications like the AWS SysOps Administrator Associate certification exam and looking for some challenging practice tests and mock tests then you have come...

Keywords: aws exam, certifications aws, aws certifications, exam aws, aws certification, aws courses, aws sysops, exam certification, aws certified, sysops exams
Source: sqlrevisited.com

Linux, Git, Docker: Why That Order Matters

Linux, Git, Docker: Why That Order Matters Every roadmap gives you the same list. Linux, Git, Docker, then a cloud, then Kubernetes, then something with the word "observability" in it. The lists are n...

Keywords: git docker, docker linux, docker underwhelming, docker commands, practice docker, learn docker, docker containers, docker obvious, linux containers, docker
Source: devops-daily.com

How Delphi ships 100 times a day with its Python backend on Vercel

Copy link to headingDelphi on Vercel 10 engineers with no dedicated infrastructure role Everyone ships code, including product and design 100+ production deploys a day behind feature flags Delphi buil...

Keywords: delphi backend, delphi builds, create delphi, delphi engineers, using delphi, delphi platform, engineer delphi, delphi worked, delphi make, backend vercel
Source: vercel.com

Issue #035 - HolmesGPT, Job retries, and the IfNotPresent hole from 2015

Capsule's API description says a tenant quota's hard limit "is never crossed." The controller behind it sums usage on a reconcile loop and rewrites each namespace's limit to whatever headroom is left,...

Keywords: ledger crashlooping, crashlooping queries, pods failurepolicy, cause robusta, crashlooping, robusta started, robusta dev, pod failure, bot robusta, robusta
Source: podostack.com

OpenCode vs Codex (2026): Which AI Coding Agent to Use

OpenCode vs Codex (2026): Which AI Coding Agent to Use OpenCode vs Codex compared for 2026 on models, surfaces, LSP feedback, openness, safety, and pricing. See which AI coding agent fits your workflo...

Keywords: agent opencode, opencode vs, openai codex, codex openai, opencode openai, vs opencode, opencode codex, openai coding, models opencode, openai built
Source: nimbalyst.com

Where Your Code Runs When There Is No Region

Where Your Code Runs When There Is No Region Pick a region. Every deployment tool you have used starts there, and the choice carries a whole model with it: a machine, or something shaped like one, run...

Keywords: edge runtimes, edge runtime, node compatibility, cloudflare workers, deploy edge, service latency, reliable cloud, edge service, latency argument, latency
Source: devops-daily.com

AI Sovereignty: Who Controls Your Company's Intelligence?

For the first few years of generative AI, the security conversation inside companies was relatively easy to understand. An employee opened ChatGPT in a browser, copied a piece of text into it and aske...

Keywords: ai conversations, security conversations, ai security, security conversation, conversation ai, discussion security, ai uses, ai maintaining, internal assistant, conversation privacy
Source: helmcode.com

Inside OpenAI’s agentic software factory

It’s rare to work with an unlimited token budget, but at OpenAI, that’s what all engineers, researchers, finance colleagues, and marketing folks do. Recently, I visited one of the world’s leading fron...

Keywords: openai engineers, engineers openai, engineering openai, openai employees, development openai, openai dev, maintained openai, openai operates, budget openai, openai headquarters
Source: newsletter.pragmaticengineer.com

Swift 6.4 Released

Swift 6.4 Released Swift 6.4 is now available. Swift aims to be a great choice across the stack, from apps and servers to systems code, embedded devices, and the browser. This release deepens that sup...

Keywords: swift platform, uses swift, swift vs, swift runtime, embedded swift, support swift, use swift, swift microcontrollers, using swift, directly swift
Source: swift.org

We got admin access to Baseten's production GitHub in 25 minutes

We were about to trust Baseten with our own and our customers’ data. So to be safe, we ran Strix to ensure they were secure first. About 25 minutes later, it had a live GitHub token with repository-le...

Keywords: baseten security, basetenbot credential, trust baseten, github token, vulnerability baseten, baseten repos, source security, github_token, membership github, token git
Source: strix.ai

Thread Trace Part 1: ROCprof Compute Viewer

Thread Trace Part 1: ROCprof Compute Viewer# Once you have identified a GPU kernel to optimize, you need to understand how its instructions execute and where waves spend time waiting. Thread trace giv...

Keywords: rocprofv3 thread, rocprof trace, rocprofv3 analyze, trace rocprof, traces rocprofv3, trace rocprofv3, analyze rocprof, thread trace, detailed tracing, profile rocprofv3
Source: rocm.blogs.amd.com

Our Software Company Now Has Hardware: Running Local AI in Production

-5b301f10ddcd---4 crawled_date: 2026-09-15T19:34:58.582326+00:00 feed_url: https://itnext.io/feed published: Tue, 15 Sep 2026 18:57:35 GMT

Three GPUs on a shelf run the automatic refresh b...

Keywords: waste heat, benchmarking energy, task engines, thermodynamics enjoy, extraction workloads, analyse unstructured, unstructured, processes billions, physical component, cooling
Source: itnext.io

Introducing ClickHouse's new TimeSeries Engine: Your drop-In Prometheus replacement

TL;DR: Store and query Prometheus metrics in ClickHouse Cloud without rewriting the PromQL queries your team already relies on. The private preview lets you send metrics via Prometheus remote write, t...

Keywords: metrics promql, promql timeseries, query prometheus, database prometheus, promql_database prometheus, promql_table metrics, prometheus data, prometheus promql_table, supporting promql, prometheus api
Source: clickhouse.com

Introducing d-Matrix Demo Cloud: Ultra Low Latency Inference, Ready to Test in Minutes

What if you could see exactly what d-Matrix Corsair does for your workloads before you ever pick up the phone with a sales team? That’s the idea behind d-Matrix Demo Cloud: a hosted environment built...

Keywords: cloud inference, matrix memory, matrix corsair, performance corsair, performance explore, inference accelerators, inference gpu, perform workloads, inference corsair, corsair servers
Source: d-matrix.ai

Your Agent Aced the Task. Will It Do It Again?

Your Agent Aced the Task. Will It Do It Again? That is embarrassing onstage. In production, it is a reliability problem: a workflow that succeeded once may fail the next time a user makes the same req...

Keywords: inconsistent tasks, reliable agent, tasks agent, introduces consistency, task agent, tool consistency, agent succeeds, agent evaluation, know consistency, generating consistency
Source: huggingface.co

How NVIDIA NVLink 6 Delivers Multi-Layer Resiliency for AI Factories

For operators of large-scale AI factories, maximizing continuous output is essential for productivity. In massive-scale AI training, every GPU in the cluster must synchronize gradients across thousand...

Keywords: goodput degradation, network compute, gpu cluster, ai workloads, ai clusters, nvidia ai, cluster productivity, workloads ai, ai infrastructure, infrastructure nvidia
Source: developer.nvidia.com

Show HN: Ordewell – turn one goal into an ordered plan of coding-agent tasks

Turn one goal into an ordered plan of coding-agent tasks — each with its own runner, model and mode — then execute and verify it.

A plan you can rewrite before a token is spent. The plan is a typed...
Keywords: plan tasks, plan task, planners task, planner taskorchestrator, task plan, planner model, planner agent, planning workflows, agent tasks, model planner
Source: github.com

How AI tool calling works (40 lines of vanilla JavaScript)

Lesson 2: tool calling - how a chat model gets hands Twenty lines of code turn a chat model into an agent. You'll have those twenty lines running today, in JavaScript and in Python, and you'll know wh...

Keywords: tool calling, loop agent, agent code, chat model, loop messages, lesson loops, loop lesson, calling chat, lesson tool, agent model
Source: buttercup.sh

Red Hat Developer Hub: Preventing compliance violations with AI coding agents

AI coding agents can generate entire services, reason about architectures, and scaffold applications in minutes. But ask one to build a service for your organization and you'll quickly notice the gap:...

Keywords: agent developer, ai agents, agentic software, coding agents, ai agent, developer agent, documented agents, code agent, agents need, agentic repository
Source: developers.redhat.com

Scaling Agents in Healthcare & Life Sciences: Lessons from Madrigal Pharmaceuticals, Abridge, and Vizient

Agent programs in healthcare and life sciences are being built under a different set of constraints than those in most industries. There’s plenty of upside if the constraints can be resolved. Success...

Keywords: agents workflows, agent clinical, individual agents, agent platform, agents work, creating agents, agent platforms, validate agents, agent tool, agents clinical
Source: langchain.com

NVIDIA CUDA-Q Logical Debuts With a 7x Fermilab Speedup and a 10x Cut in Diraq’s Qubit Estimate

NVIDIA has added CUDA-Q Logical to its open-source CUDA-Q platform, an orchestration layer for building applications that run on fault-tolerant quantum computers, and it arrives with two numbers that...

Keywords: quantum gpu, quantum nvidia, nvidia quantum, qubit hardware, qubits researchers, qubit processor, quantum computer, supercomputers quantum, quantum computers, quantum computing
Source: storagereview.com

The Inference Hardware Revolution of 2026

The AI Inference Revolution Is Here Today’s tidal wave of queries is forcing hardware makers to pivot Since about 2020, AI has largely focused on training bigger and better models. Large language mode...

Keywords: 2020 ai, inference revolution, inference innovation, ai currently, current ai, improve inference, ai largely, ai training, inference performance, does ai
Source: spectrum.ieee.org

Closing the IPv6 First-Packet Gap with Grand

There is a subtle asymmetry in IPv6 Neighbour Discovery: a host knows how to reach its router before the router necessarily knows how to reach the host. GRAND fixes that asymmetry by making the host p...

Keywords: ipv6 neighbour, ipv6 hosts, asymmetry ipv6, host ipv6, transmitted ipv6, knows ipv6, situations ipv6, announces ipv6, ipv6 packet, cause ipv6
Source: labs.ripe.net

How much of F-Droid is LLM generated?

I love F-Droid. I love what F-Droid stands for and I like the freedom that it gives its users. As a FOSS app maintainer I also have nothing but good things to say about the people behind the project....

Keywords: droid ai, droid, thinking droid, ai apps, droid just, love droid, droid love, ai smells, open droid, ai suspicious
Source: tintotint.eu

Show HN: Panel – A research workspace where the agent can build its own panes

A research workspace where the agent works beside you: chat, files, PDFs and notebooks in one dock, and the agent can also create custom viewers and apps when necessary This is an early build for test...

Keywords: workspace agent, dock agent, pnpm dev, run pnpm, sync pnpm, chatting agent, folder agent, workspace specifically, agent create, pnpm install
Source: github.com

1Password's AI patching benchmark is misleading

1Password's AI patching benchmark is misleading 1Password’s FLAWED report, published on August 6, 2026, gives defenders a misleading picture of AI patching. Its headline says models produced clean fix...

Keywords: ai patching, agent patch, patches agent, test patches, patching benchmark, patches report, code patch, improve patches, patch authored, vulnerable patched
Source: blog.trailofbits.com

Windows 11 KB5129195 is out after Microsoft confirms major issues with the September 2026 update, but it won’t fix AMD GPU errors

Windows 11 KB5129195 is rolling out now, and it addresses multiple critical issues we found in our tests, including a bug that broke audio on some PCs. Microsoft told me today’s emergency update downl...

Keywords: kb5124008 patch, kb5124008 bug, update kb5129195, vulnerability kb5129195, update kb5124008, kb5129195 fixes, kb5129195 emergency, kb5129195 rolling, kb5124008 september, windows updates
Source: windowslatest.com

Dense vs. MoE Models: Active Parameters, Throughput, and When to Choose Each

How can a 30B-parameter model activate only 3B parameters per token, and still use the capacity of the larger model? Nemotron 3.5 Lightning illustrates the answer: It uses a Mixture-of-Experts (MoE) a...

Keywords: models faster, 27b parameters, model performance, model nemotron, parameter model, model faster, model architectures, manageable parameters, complexity nemotron, 30b parameter
Source: developer.nvidia.com

Looking at High-Reliability Compute and Storage

Looking at High-Reliability Compute and Storage What you’ll learn:

What technologies do Penguin Solutions specialize in?
How has high-reliability, data center compute changed over time?
The evol...
Keywords: penguin computing, technologies penguin, storage penguin, storage devices, flash storage, compute storage, high reliability, dram sd, embedded memory, cxl storage
Source: electronicdesign.com

Siri AI 📱, Zuckerberg profile 🧍‍♂️, Xcode 27 👨‍💻

TLDR 2026-09-15 Siri AI 📱, Zuckerberg profile 🧍♂️, Xcode 27 👨💻 The AI employee 60,000+ teams have hired since March (Sponsor) A PM posts in #product: "Can we get a churn dashboard, refreshed daily?" V...

Keywords: tasks siri, capabilities siri, ai employee, models siri, apple siri, apple intelligence, improved siri, siri extension, siri model, ai job
Source: tldr.tech

Small Programming Tricks

Small Programming Tricks Day to day, I think a surprising amount of engineering productivity comes from small nuggets of knowledge: being aware that a language feature exists; knowing that an unexplai...

Keywords: know python, python3 http, use python3, python3, python use, small programming, python, simple server, tcp_no_delay, tcp delay
Source: will-keleher.com

The 3 Layers of Testing AI Agents

The first time someone on the team asks: "How do we test the agent?", the room splits into two wrong answers. One camp says you can't, it's non-deterministic, we'll watch it carefully in production. T...

Keywords: agent testing, mocks llm, mocks agent, testing agent, agent test, test agent, tests llm, camp mocks, assertion agent, mocks
Source: thetshaped.dev

Alternatives to MinIO for single-node local S3

In late 2025 the company behind MinIO decided to abandon it to pursue other commercial interests. As well as upsetting a bunch of folk, it also put the cat amongst the pigeons of many software demos t...

Keywords: s3 minio, minio s3, minio alternatives, alternatives minio, minio demos, minio promising, emulate s3, minio emulate, emulating s3, using minio
Source: rmoff.net

Build an interview coach app with the GitHub Copilot SDK

An interview coach has to do more than ask questions. It needs to read a resume, follow up on an incomplete answer, and save enough context to give useful feedback at the end. Some of that work is con...

Keywords: copilot interview, app interview, interview tools, interview coach, interview needs, interviewer technical, interviewdata uses, agent interview, azure model, tools interview
Source: devblogs.microsoft.com

Codex vs Cursor (2026): Which One Should You Use?

Codex vs Cursor (2026): Which One Should You Use? Codex vs Cursor compared for 2026 on agent autonomy, editing, models, surfaces, parallel sessions, code review, pricing, and open source. A clear, cur...

Keywords: codex compared, codex vs, pick codex, codex better, codex cursor, tool codex, editor codex, cursor codex, better codex, codex handling
Source: nimbalyst.com

[email protected]

78aadc8: Bundle Mermaid and the EPUB generator only for sites that use them. Both were dynamic imports, so they never loaded on a page that didn’t need them, but they were still built on every blume b...

Keywords: mermaid epub, sidebar sections, sidebar groups, bundle mermaid, bundling sites, page mermaid, sidebar group, site sidebar, carries sidebar, generator sites
Source: useblume.dev

Scaling Telco Autonomy: Leveraging GNNs with Distributed GraphFlow

Scaling Telco Autonomy: Leveraging GNNs with Distributed GraphFlow Brian Naughton Senior Principal Architect, Telecommunications Mathieu Guillame-Bert Software Engineer The telecommunications industry...

Keywords: network ai, leveraging gnns, networks gnns, scalable gnn, gnn modeling, intelligent network, graph neural, using gnns, vertex ai, gnn expertise
Source: cloud.google.com

Best practices for handling cloud reliability incidents

Best practices for handling cloud reliability incidents Flemming Christensen Product Manager & Technical Solutions Engineer, Google Cloud Cloud outages can range from global service disruptions to iss...

Keywords: cloud reliability, cloud outages, handling cloud, outage impacting, resolving outage, resolve outage, behavior outage, reliability incidents, cloud support, impact outage
Source: cloud.google.com

Fedora 45 Beta Released With KMSCON Console, GNOME 51 & Other Updates

Fedora 45 Beta Released With KMSCON Console, GNOME 51 & Other Updates The beta release of Fedora 45 is out today and right on schedule for this wonderful Linux distribution. Fedora 45 is another featu...

Keywords: fedora 45, release fedora, updates fedora, fedora workstation, desktop fedora, announcement fedora, reproducible fedora, today fedora, gnome 51, distribution fedora
Source: phoronix.com

GDScript: The Good, Bad, and Ugly Parts

I recently finished writing a fairly significant amount of code—implementing a system first in TypeScript, and then porting that same system to GDScript so it can run on Godot (I have a reason to need...

Keywords: gdscript language, gdscript useful, gdscript designed, does gdscript, gdscript support, gdscript programming, gdscript really, gdscript conception, gdscript nature, gdscript instead
Source: azhdarchid.com

Jev: New frontier model 40-400x cheaper and 20-200x faster

Company News Sep 14, 2026 Introducing System One Models and Jev Diogo Almeida, founder, TypeSafe Models have been superhuman at chat for years, so where is all the automation? This has been my driving...

Keywords: chat models, existing models, introducing models, typesafe ai, models jev, model capabilities, ai releasing, better models, research chatgpt, focused automation
Source: typesafe.ai

Micron Shows off 512GB DDR5 RDIMM: 12TB per Dual-Socket Server at 9,200 MT/s, Volume Production in 2H 2027

Micron has demonstrated a 512GB DDR5 RDIMM running on multiple server platforms, which it calls the world’s first module at that capacity, and says AMD and Intel are both validating it for their next-...

Keywords: micron 512gb, performance micron, micron ddr5, 12tb micron, server memory, 512gb ddr5, 128gb rdimms, capacity memory, stacks dram, 512gb rdimm
Source: storagereview.com

Show HN: Capsule – Single-file web apps that save their data into SQLite

Documents that run like apps Capsule packs your entire app — UI, data, and everything — into a single portable .capsule file. No cloud. No accounts. Just share it. Generate & update apps using AI Caps...

Keywords: apps capsule, app capsule, portable apps, applications cloud, app create, setup apps, building app, file capsule, capsule files, capsule file
Source: withcapsule.app

Sound Open Firmware 2.15 Released With AMD ACP 7.x Support, Intel UAOL

Sound Open Firmware 2.15 Released With AMD ACP 7.x Support, Intel UAOL Sound Open Firmware as the open-source, vendor-independent audio DSP firmware stack and driver framework is out today with a new...

Keywords: amd acp, upcoming amd, audio processor, including amd, acp audio, dsp firmware, firmware stack, amd zen, released amd, audio dsp
Source: phoronix.com

Tell agents the why, not just the how

Tell agents the why, not just the how Early AI agents were basically enthusiastic idiots. Working with them required you to tell them precisely what you wanted them to do (for instance, “method A exis...

Keywords: tell agents, prompting agents, agent context, ai agents, agents just, agents improved, agents agent, agents basically, agents, early ai
Source: seangoedecke.com

The READ METADATA privilege will soon be available by default for workspaces with the compliance security profile enabled

What's coming? Learn about features and behavioral changes in upcoming Databricks releases. The READ METADATA privilege will soon be available by default for workspaces with the compliance security pr...

Keywords: metadata genie, role databricks, agents databricks, authorization databricks, databricks governance, metadata privilege, access databricks, manage databricks, enabled databricks, workspaces databricks
Source: docs.databricks.com

What I learned organizing KCD Lima 2026

On July 18, 2026, we held the third edition of Kubernetes Community Days Lima at UTEC in Barranco. By now, we have already sent the Transparency Report to the CNCF, thanked our sponsors, and processed...

Keywords: kubernetes community, lima organizers, awaited kubernetes, edition kubernetes, kubernetes, kcd organizers, kcd organizing, kcd lima, lima devopsdays, devopsdays lima
Source: cncf.io

Coder Brings Claude Code to Agent Relay, Unlocking Agentic Development for the World’s Most Regulated Enterprises

Coder’s Agent Relay integration with Claude Code, built with support from the Claude Code team, lets financial services, technology leaders, and global enterprises run the industry’s leading coding ag...

Keywords: coder agent, coding agent, code agents, coding agents, code agentic, agent infrastructure, cloud agents, infrastructure coder, agent provider, agent relay
Source: coder.com

Handling non-Latin fonts, standard fonts, and JPEG 2000 in react-pdf

Handling non-Latin fonts, standard fonts, and JPEG 2000 in react-pdf Table of contents react-pdf for special PDF requirements — CMap files for Chinese, Japanese, and Korean text; standard font data fo...

Keywords: react pdf, pdfjs, pdfjs dist, pdfs use, characters pdfs, pdf js, fonts pdf, wasm pdfs, pdfs chinese, standardfontdataurl pdf
Source: nutrient.io

How NVIDIA Groq 3 LPX Deterministic Execution Drives Power-Efficient High-Interactivity Inference on NVIDIA Vera Rubin

Power is a defining constraint for AI factories. As AI workloads demand a full compute platform to serve them, each component of that platform must maximize output within the factory’s limited power b...

Keywords: nvidia vera, technologies nvidia, powering ai, level nvidia, power efficient, core nvidia, generation nvidia, nvidia, power provision, power ai
Source: developer.nvidia.com

Optimistic vs Pessimistic Concurrency in .NET

Almost everyone had this situation in production: a customer changed the delivery address on a shipment, saved it, saw the confirmation, and an hour later the old address is back. You start investigat...

Keywords: dbupdateconcurrencyexception retrying, disappearing concurrency, handling dbupdateconcurrencyexception, test concurrent, catch dbupdateconcurrencyexception, handle dbupdateconcurrencyexception, throws dbupdateconcurrencyexception, dbupdateconcurrencyexception dbupdateconcurrencyexception, dbupdateconcurrencyexception maxretryattempts, failures retry
Source: antondevtips.com

Rail Network Modernization: FRMCS, Sovereignty and the AI Edge

A pre-InnoTrans 2026 perspective. https://www.innotrans.de Cisco will be at InnoTrans 2026 in Berlin from 22–25 September, at Messe Berlin — Hall 4.1, Booth 450 — as rail infrastructure managers and o...

Keywords: gsm frmcs, railway gsm, communications railway, future railway, railway mobile, phased gsm, infrastructure frmcs, gsm global, network frmcs, cisco innotrans
Source: blogs.cisco.com

Swift 6.4 makes Swift Build the default and speeds browser bridging

Swift 6.4 makes Swift Build the default and speeds browser bridging The Apple-backed project also stabilizes Subprocess and expands interoperability with C++, Java, Android and embedded targets. By Ru...

Keywords: build swift, built swift, swift build, swift compiler, swift package, introducing swift, make swift, swift released, extension swift, standard swift
Source: runtimewire.com

From Megawatts to Tokens: How NVIDIA Maximizes AI Factory Production

On a sweltering August evening in Silicon Valley, as the sun dropped and air conditioning loads spiked, Silicon Valley Power sent a signal to an AI factory to adjust its power consumption. Varun Sivar...

Keywords: nvidia ai, ai workloads, valley power, ai workload, gpu cloud, silicon valley, software power, accelerated computing, electricity gpus, ai san
Source: blogs.nvidia.com

How I Wrote a Forth (Without Knowing How)

For the past few weeks, I have been writing an implementation of the Forth programming language. I had not written a Forth before, and did not really know how to write a Forth before setting out to wr...

Keywords: forth programming, forth process, writing forth, forths implement, write forths, forth implementing, implement forth, write forth, construct forth, forth written
Source: vtrlx.ca

The Bitter Lesson of Browser Agents

When we launched Browser Use in November 2024, GPT-4o wasn't trained for computer use. Built on next-token prediction, it didn't reliably understand what it could do on a page. We had to spell it out....

Keywords: actions browsers, browser automation, browser state, browser tools, browser tool, state browser, commands browser, harness browsercode, tools browser_exec, actions code
Source: browser-use.com

A Nix store is three functions

While building trynix I needed somewhere to host a store-path that did not exist on cache.nixos.org.11I was also waiting for @domenkozar to enable CORS on cache.nixos.org so I could use it. I wanted t...

Keywords: fetched nix, cache nixos, nix cache, storepath nix, nix fetching, nixpkgs nixos, nixpkgs store, store nix, non nixpkgs, nix archives
Source: fzakaria.com

I Tried 20+ CouchDB Courses - Here Are My Top 5 Recommendations for 2027

Many people are going gaga over CouchDB. But what exactly is it? And why should you care? CouchDB is what I would call one of many NoSQL solutions. It is a shift away from what is called traditional d...

Keywords: differences couchdb, difference couchdb, databases couchdb, couchdb database, apache couchdb, database couchdb, languages couchdb, couchdb documented, concepts couchdb, couchdb nosql
Source: sqlrevisited.com

Trusted Answers at Enterprise Scale

Trusted Answers at Enterprise Scale Dell AI Solutions with Open Models Authors: Brian Martin September 15, 2026 Enterprise AI has moved from pilots to production workflows, where its most common job i...

Keywords: trusted ai, enterprise ai, answers enterprise, answers trusted, trusted answers, enterprises agentic, reliability ai, ai retrieves, ai solutions, agentic readiness
Source: signal65.com

Why Personal Websites Are Coming Back

For a while, the personal website looked like a leftover from an earlier Internet. In the 1990s and early 2000s, having your own home page was fairly normal for anyone who spent much time online. Inte...

Keywords: personal website, personal websites, websites personal, websites accounts, website business, personal sites, maintaining website, personal site, websites technology, sites depended
Source: deadparrotbbs.com

Daily Reading List – September 14, 2026 (#866)

I’m in New York City for a day and had fun judging a hackathon conducted by Major League Baseball. AI helped all sorts of people build all sorts of things. In between all that, I walked seven miles ar...

Keywords: baseball ai, ai, ai development, feature flags, ai personal, ai agents, using ai, skeptical ai, feature flag, safety net
Source: seroter.com

WangNet – 1.8 MB, zero-dependency Numberwang adjudication in 11 languages

A small neural network that decides whether a number is Numberwang. The whole model is a 1.8 MB JSON file and the inference code is about 100 lines of pure Python standard library — no PyTorch, no Num...

Keywords: numberwang python3, python3 numberwang, numberwang py, numberwang confidence, playing numberwang, p_not_numberwang, numberwang import, numberwang model, p_not_numberwang p_numberwang, p_numberwang
Source: github.com

Agent of the Day – September 15, 2026

Agent of the Day – September 15, 2026 Custom lint rules are a strange kind of code: they’re supposed to catch other people’s mistakes, but nobody is watching them for their own. A rule ships, looks re...

Keywords: eslint rules, refiner eslint, eslint refiner, linter eslint, custom eslint, eslint factory, day eslint, eslint, rules agent, rules lint
Source: github.github.com

GEFS on OpenBSD: A very early preview

[prev in list] [next in list] [prev in thread] [next in thread] List: openbsd-tech Subject: GEFS on OpenBSD: A very early preview From: ori () eigenstate ! org Date: 2026-09-15 15:45:05 Message-ID: EB...

Keywords: gefs openbsd, gefs running, gefs new, gefs gefs, gefs, gefs note, talk gefs, dev gefs, fetch gefs, gefs diff
Source: marc.info

How to Prepare for Coding Interviews in 2026? (with Resources)

Hello guys, are you preparing for coding interviews but not sure where to start with? Are you drowned in the sea of Data structures and algorithms and System design and what not then you have come to...

Keywords: coding interview, preparing coding, coding interviews, coding beginner, prepare coding, interviews prepare, programming, design interview, start interview, courses prepare
Source: java67.com

CSS-Tricks in Limbo

CSS-Tricks in Limbo I’m sad to say that CSS-Tricks is stuck in limbo again. The site was acquired by Digital Ocean in 2022 and it continued to run under their ownership until February of 2023 when D...
Keywords: digitalocean ceo, acquired digital, ocean hired, announcing digitalocean, digital ocean, digitalocean funding, digital, digitalocean ceased, 2023 digital, digitalocean pledged
Source: vale.rocks

E2B reintroduces its open infrastructure as E2B Runtime

E2B on X: "🚀 Our open-source project has a new name: E2B Runtime. Behind every E2B stack there's a runtime. Cloud, enterprise, or embedded in your product. Same code, Apache-2.0. Old URL redirects. So...

Keywords: new e2b, e2b runtime, runtime e2b, hosted reference, e2b, e2b open, apache old, e2b stack, apache keeping, redirects new
Source: x.com

Perplexity says persistent agents helped two engineers replace DynamoDB

We built a replacement for AWS DynamoDB, a key-value database for fast web content fetches. This was done with two engineers and hundreds of persistent Computer agents over two months. Migrating to ou...

Keywords: dynamodb, aws dynamodb, dynamodb key, database fast, replacement aws, aws, fast web, house database, computer agents, search engineers
Source: x.com

Shadowing the Standard Library

I was reading Johann Rehberger’s write-up of an attack on a coding agent last week. The payload is a file called struct.py inside a zip archive. The agent unpacks the archive, writes its own small Pyt...

Keywords: python resolves, poisoned module, interpreter runs, module level, python behaviour, struct python, pythonsafepath, module path, module python, import struct
Source: nesbitt.io

Why I'm still bearish on LLMs after Navier-Stokes

why i'm still bearish on LLMs after navier-stokes [thank you to claude fable 5.1, holden saberhagen, gabriel kammer, andres erbsen, alice mckean, and tristan wylde-larue for comments on this post] i'l...

Keywords: caveats frontier, frontier models, frontier labs, propose frontier, models supervise, current frontier, cheap frontier, frontier lab, use frontier, demanding expertise
Source: dank.systems

Inside Google’s Database Infrastructure for the AI Era

Inside Google’s Database Infrastructure for the AI Era Historically, databases were responsible for storing data and returning exact results in response to queries. However, AI is now bending that con...

Keywords: google database, engineering google, databases evolved, queries ai, databases, queries propose, database infrastructure, databases responsible, softwareengineeringdaily, databases territory
Source: softwareengineeringdaily.com

LLMs Have a New Limit – It Costs More to Think Longer

NVIDIA’s Igor Dmochowski explained that LLMs are no longer limited just by model size or benchmark scores; the cost of longer reasoning means runtime, context length, memory use, and throughput mat...

Keywords: llms longer, explained llms, llms new, llms, dev llms, cost longer, longer limited, length memory, size benchmark, nvidia
Source: shiftmag.dev

Mayfly Chat: Transient Chat for Agents

Earlier this week I caught myself copy/pasting code between two agents. I don’t want to be a meat proxy, but it really was going to be best if they just…talked to each other. And yet it was weirdly ha...

Keywords: agents chat, chat agents, transient chat, mayfly chat, communicating mayfly, agent vm, agents create, forwarding agent, summoned mayfly, ask agents
Source: blog.exe.dev

The broken windows theory of coding agents

The broken windows theory of coding agents When my most experienced engineer got tired of our shitty code Thanks Linear for being a long-term supporter of manager.dev! It’s the product my team enjoys...

Keywords: agent review, code agents, coding agents, coding agent, agents experienced, code reviewers, code reviews, agents write, reviewing code, understand agents
Source: newsletter.manager.dev

Linux from Scratch

Welcome to Linux From Scratch! Linux From Scratch (LFS) is a project that provides you with step-by-step instructions for building your own custom Linux system, entirely from source code. Currently, t...

Keywords: lfs linux, lfs development, custom linux, multilib linux, blfs linux, lfs project, describes lfs, useful lfs, supplemental linux, lfs ways
Source: linuxfromscratch.org

Planning with Agents: Divided Worlds, Boundary Objects, and Thicker Interfaces

This talk is about collaborative planning with agents: divided worlds, boundary objects, and thicker interfaces But let’s first talk about Thomas Gladwin , an anthropologist who spent time between the...

Keywords: planning agents, collaborative planning, planning agent, planning communicating, planning medium, thinking planning, planning, interactive plans, plans orient, navigator approach
Source: maggieappleton.com

Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations

E-ink bird frame for Raspberry Pi - real-time bird detection by audio, fully local AI, rendered as real, hand-cut 1800s bird illustrations. Sorry about the dirty window - squirrels have been stealing...

Keywords: detection birdnet, birdnet listens, run birdnet, bird illustrations, bird images, bird detection, birds images, birdnet, birdnet live, birds detected
Source: github.com

Stalling installing

I’m an invited expert at the World Wide Web Consortium. That sounds impressive, but it isn’t. Anyone can become an invited expert. The fact that I am now one proves it. You apply to be an invited expe...

Keywords: web consortium, apps webkit, webkit concerned, expert web, webkit rejecting, web apps, expert application, installable web, webkit stance, install web
Source: adactio.com

VQ-bench: A Composable Vector Quantization Framework

Abstract Vector quantization is an old problem but has recently become central to AI infrastructure. It is therefore experiencing a surge of renewed engineering and research activity. This paper provi...

Keywords: vector quantization, quantization algorithms, new quantization, quantization, quantizers pipelines, common quantizers, quantization old, quantization primitives, conceptual quantization, quantizers
Source: pinecone.io

Beijing Paces at the Gate

FlashPoint is our premium quick-strike column on market-moving China tech events. Today: Anthropic CEO Dario Amodei published a pacing proposal on September 12, drawing public support from Sam Altman...

Keywords: china deployment, china filing, question china, moving china, china foreign, china tech, mapped china, china national, beijing moving, beijing
Source: hellochinatech.com

Bypassing inference bottlenecks: Accelerating complex AI search with Retrieve-for-Train

Algorithms & Theory

Keywords: algorithms theory, algorithms, theory
Source: research.google

JDK 27 Security Enhancements

JDK 27 was released on September 15, 2026! As with my previous blogs, I have compiled a list of what I think are the most interesting and useful security enhancements in this release. I have also grou...

Keywords: jdk tls, tls jdk, improvements jdk, introduced jdk, java cryptographic, jdk 27, jdk releases, jdk security, features jce, thejdk tls
Source: seanjmullan.org

MutexaGPT turns enzyme design ideas into physics simulation workflows

MutexaGPT: An Intuition-to-Design Translator for Physics-Based Enzyme Engineering @NatComputSci

The paper presents MutexaGPT, an open-access, multi-agent LLM platform that turns plain-English enzym...

Keywords: proteinengineering, enzymeengineering proteinengineering, enzymeengineering, enzyme engineering, proteinengineering moleculardynamics, biocatalysis highthroughputcomputing, protein engineering, s43588 enzymeengineering, moleculardynamics computationalbiology, molecular simulations
Source: x.com

Treating Prompt Templates as Hyperparameters in Scikit-LLM GridSearchCV

In this article, you will learn how to treat prompt templates as tunable hyperparameters for a language model, using scikit-learn’s grid search to find the best-performing prompt for a zero-shot text...

Keywords: classifier hyperparameter, hyperparameters scikit, hyperparameters language, hyperparameters words, tunable hyperparameters, hyperparameters, hyperparameter optimization, hyperparameters machine, hyperparameter, classify text
Source: machinelearningmastery.com

ModelScope releases an 8B mixture model that activates 1.2B parameters for local inference

Only ~1.2B parameters active at a time. Edge0-8B-A1B-preview makes an 8B-class MoE practical for local inference.📜 Apache 2.0. 🤖 modelscope.ai/models/Edge0/E… ⚡ Reaches 23.9–25.3 tokens/s with about 1...

Keywords: benchmarks mmlu, models edge0, benchmarks, edge0 8b, benchmark, memory prerouter, benchmark retains, context benchmark, mmlu pro, active memory
Source: x.com

Show HN: Sunk Cost – How long until a local LLM rig pays for itself?

The break-even How you’d use it Assumptions you can change Where nothing has been measured, local speed is estimated as memory bandwidth ÷ bytes read per token × , and labelled as such. API speed only...

Keywords: api speed, local speed, speed estimated, speed affects, memory bandwidth, memory measured, bandwidth bytes, time comparison, measured local, speed
Source: sunkcost.ai

Simple approximation for spherical cap area

The previous post looked at how to interpret cosine similarity, or equivalently angles between word vectors. In a high-dimensional space, randomly chosen vectors are likely nearly perpendicular, and s...

Keywords: cosine similarity, spherical cap, similarity cos, interpret cosine, cos approximation, dimensions spherical, sphere moderately, spherical, cap_fraction_approx theta, cosine
Source: johndcook.com

When the Test Broke Containment

Key Takeaways

Roughly 1,200 supposedly isolated agents communicated through an unauthorized message board, and about 700 participated in Hugging Face-related activity.
Agents exchanged over 70,000...
Keywords: agents spoofing, security incident, agent forensic, face security, agents communicated, investigators agents, isolated agents, agents deceptive, agents door, agents bypassed
Source: commvault.com

Java 27 Released

JDK 27, the reference implementation of Java 27, is now Generally Available. We shipped build 35 as the second Release Candidate of JDK 27 on 20 August, and no P1 bugs have been reported since then. B...

Keywords: jdk 27, java 27, jdk java, openjdk builds, jdk, projects jdk, openjdk, https jdk, openjdk org, licensed openjdk
Source: mail.openjdk.org

Postgres development activity

Postgres development activity Every now and then I need a break from writing code. In those cases I like looking at data about a subject I’m interested in - looking for trends, quantifying the expecte...

Keywords: postgres development, postgres project, development activity, git history, commit stats, project pgsql, discussed git, participating development, postgres, pgsql hackers
Source: vondra.me

Single Digit Thousand Dollar AI SDR

Vercel took its inbound sales development team from 10 people to 1.25. Jeanne DeWitt Grosser, COO of Vercel, shared this on our Information interview together1 : “We had 90% automation of sales develo...

Keywords: ai early, sales development, development inbound, getting ai, agent ve, automation sales, inbound sales, development agent, today ai, inbound agent
Source: tomtunguz.com

The terminal should not own the work

The terminal should not own the work A terminal is a good place to start a process. It is a poor place to store the truth about that process. When an agentic task belongs to one terminal tab, the tab...

Keywords: terminal work, work terminal, daemon owns, reopening project, owns task, terminal good, agent processes, task close, vincent persists, handled vincent
Source: lezli01.is-a.dev

GenCDR generates antibody binding loops from framework sequences

Generative language modeling for antibody CDR grafting and alignment-driven de novo design

The paper introduces GenCDR, a framework-first autoregressive (LLaMA-style) antibody language model: it re...

Keywords: antibody cdr, modeling antibody, novo antibody, antibody language, cdr1 novo, alignment cdr, antigen encoding, gencdr framework, introduces gencdr, novo generation
Source: x.com

LM Studio adds local real-time voice transcription across Mac, Windows, and Linux

LM Studio on X: "Introducing local realtime voice transcription. Speak freely to your agents, without sending your voice data to the cloud. Available for Mac, Windows - and now also Linux!" Introducin...

Keywords: realtime voice, voice transcription, lm studio, transcription speak, voice data, local realtime, sending voice, voice, realtime, transcription
Source: x.com

Only 7% of Organizations Using Kubernetes for AI Do So Daily

Two-thirds of organizations run AI workloads on Kubernetes, but only 7% do it daily. Katie Gamanji says the ecosystem still needs better operational maturity and standards. The post <a href=...

Keywords: kubernetes ai, workloads kubernetes, kubernetes daily, using kubernetes, kubernetes, ai workloads, ai daily, organizations run, run ai, organizations using
Source: shiftmag.dev

Show HN: Check if your IP has appeared in a residential proxy network

Have I been proxied? Find out in one click. Check whether your public IP has been observed in a residential proxy network, and understand what to do next. We check the public IP address your browser i...

Keywords: check proxied, proxies, proxied, residential proxies, proxied powered, proxy, residential proxy, ip observed, proxies vpns, proxy network
Source: haveibeenproxied.com

The CSS Zen Garden dream shipped

August 14, 2026 The CSS Zen Garden dream, finally shipped Rebuilding Firefox.com with Mozilla and Lincoln Loop on modern, native CSS with no preprocessors, and what that means for design systems today...

Keywords: css zen, css preprocessors, discovered css, pure css, modern css, authored css, legacy browsers, css depends, css written, apart css
Source: josprague.com