Insta360 has launched its first gimbal camera, the $770 Luna Ultra.

Following months of teases, leaks, a secretive debut at NAB 2026, and an initial launch in China, Insta360 finally announced global availability for its first handheld stabilized camera. The Luna Ultra features a pair of 8K cameras atop a 3-axis gimbal, offering an upgrade to DJI's Osmo Pocket 4 and Pocket 4P, which both max […]

Your CI was green last Friday. Today, the payments test is failing. Somewhere between Friday's merge and now, 47 commits landed on main . Which one broke it? Most developers answer this the wrong way: they scroll through git log , check out suspicious commits one by one, and run the test manually. An hour later, they're still guessing. There's a command built for this exact problem. It's called git bisect , and once you learn it, you'll never debug regressions the old way again. How bisect works git bisect is a binary search across your commit history. You tell Git two things: A good commit (a known point where the bug didn't exist) A bad commit (a known point where the bug exists — usually HEAD ) Git then checks out the commit halfway between them. You test. You mark it as good or bad. Git narrows the range by half. Repeat. With 47 commits between "good" and "bad", it takes at most 6 steps (log₂ 47) to find the exact commit that introduced the bug. Versus checking every commit manually, that's the difference between 5 minutes and an hour. The manual workflow # Start a bisect session $ git bisect start # Mark the current state (HEAD) as bad $ git bisect bad # Mark a known-good commit $ git bisect good a3f1d22 # Bisecting: 23 revisions left to test after this (roughly 5 steps) # [7e4b9c1] refactor: extract payment validator # Git has checked out a commit in the middle. Run your test. $ npm test -- --grep "payments" # Test passed — mark this commit as good $ git bisect good # Bisecting: 11 revisions left to test after this (roughly 4 steps) # [b2d8e11] feat: add retry logic to payment API # Test failed — mark this commit as bad $ git bisect bad # ... continue until Git announces the first bad commit: # b2d8e11 is the first bad commit # commit b2d8e11 # Author: leo@company.com # Date: Tue Apr 15 11:42:03 # feat: add retry logic to payment API # Done — reset to where you started $ git bisect reset In 6 commands, you know exactly which commit broke the tests. No guessing

You built a workflow worth sharing — and it works perfectly. Until someone else imports it. The bottleneck is the API key. Use yours, and every user is billed against your account. Use theirs, and they each have to find the credential UI, paste their key, and reconnect every time. Both are friction. The cleaner option is to let the workflow ask for the key on the form, then thread it through to the HTTP nodes that need it. It's simpler than it sounds. This post walks through the pattern with a working credential setup, an alternative for single-node simple cases, the gotchas, and a note on what this enables for custom node authors. The screenshots below come from n8n's built-in Bearer Auth credential and from the n8n-nodes-ldxhub package's own credential schema. The technique itself is generic — what's shown here works for any HTTP-node workflow and any custom node that supports expression-mode credentials. The form asks, the credential listens The simplest case: a Form Trigger collects an API key, then an HTTP node hits an authenticated endpoint with that key. Two nodes, one bridge between them — but the bridge isn't a direct expression. It runs through a credential. The flow: Form Trigger collects api_key (use the Password element type for masking) A Bearer Auth credential references that form input via expression HTTP node picks the credential The Form Trigger is straightforward. Add one field: Form Trigger Form Fields : - Label : API Key - Element Type : Password - Custom Field Name : api_key - Required Field : yes Element type matters. Use Password instead of Text and the input gets masked on screen — the key isn't readable to someone glancing at the browser. Here's the rendered form a user sees when they open the workflow URL: Wiring the credential to expression mode For a Bearer token (which is what most modern APIs use), create a new credential of type Bearer Auth — a generic credential built into n8n that's purpose-built for Authorization: Bearer ... header

Uber's CTO told the world this month that the company spent its entire 2026 AI allocation by April. The story has been reported in a handful of outlets, hit the front page of Hacker News for 397 points and 469 comments , and is mostly being read as a cost-of-AI-tools story. It is one. It is also, on a closer reading of the numbers, a pricing-model story — and the structural fact that almost none of the coverage has emphasized is the one that determines whether this is a one-company anomaly or the beginning of an industry-wide budgetary crisis. The structural fact is that Claude Code, like most enterprise AI tooling in 2026, is priced on token consumption, not per-seat licensing. Token-based pricing scales with how aggressively the tool is used. Per-seat enterprise SaaS pricing — the model corporate IT budgets are built around — scales with how many people have access to it. Those two cost curves diverge in exactly the territory where productivity tools are designed to operate: high-engagement, daily-use, gradually-deepening workflows. The Uber data is the first public-facing version of a math problem most enterprise IT departments are about to discover privately. The numbers Uber CTO Praveen Neppalli Naga , named in Yahoo Finance's and Benzinga's coverage, said publicly that Uber is "back to the drawing board" on AI budgeting after the surge in Claude Code use blew through internal projections. The specific numbers, as reported across the multiple outlets covering the story: Claude Code adoption inside Uber's ~5,000-engineer organization went from 32% to 84% over four months. 70% of committed code at Uber is now AI-originated. 11% of live backend updates are "being written by AI agents built primarily with Claude Code," per the reporting. Per-engineer monthly API costs: $500 to $2,000. Uber's annual R&D spend is around $3.4 billion , of which the AI tooling line was a much larger fraction than expected. Cursor adoption plateaued; Claude Code dominated. These are ext

You’ve spent weeks, maybe months, tracking down bugs, optimizing your user interface, and wrestling with backend security rules. You compile your native release build or run your final production compilations, thinking the hardest part of the journey is officially behind you. Then you open the Google Play Console, and you’re hit with the ultimate indie developer roadblock: the mandatory 12-tester and 14-day closed testing requirement . Many independent creators view this process as a simple download checklist. You might think, "I'll just find 12 people to download the app, leave it on their phones for two weeks, and wait it out." However, treating the testing phase as a static metric is the fastest way to get rejected during the final production access review. So, what is Google actually tracking in the background during these two weeks? Let’s take a deep dive into the core algorithmic requirement that determines your success: Continuous Engagement . 🔄 Decoding "Continuous Engagement" Google Play policies are not designed as a simple box-checking exercise. The underlying goal of the algorithm is to verify if your application is genuinely functional, stable, and being tested by an organic user base before it reaches millions of production users. To enforce this, Google's advanced systems actively monitor the devices connected to your closed test track over the 14-day timeline: Background Device Pings: Google Play Services regularly collects background automated signals (ping logs) from the devices where your test build is active. Real User Interaction: Leaving an app to rot in an application drawer without ever opening it is instantly flagged by the algorithm. Google measures whether the app is actively opened daily and tracks active interaction metrics within the build. Feedback Loops: The system monitors whether your test community is utilizing the internal testing channel on the Play Store to send private developer feedback and crash reports. 📉 The Illusion of "Ju

Anthropic's strongest model is free until June 22 — and two more shifts for builders Three things landed for builders at once: the best model got cheaper (free, actually), free inference showed up on Apple's stack, and one still photo now becomes a talking video. Two of them you can act on right now. Here's the 90-second video version if you want the quick pass first: 1. Claude Fable 5 is public — and free on your plan until June 22 Anthropic released Claude Fable 5 , the first publicly available version of its Mythos-class model. It's state-of-the-art on nearly every benchmark Anthropic tests — software engineering, knowledge work, vision, and scientific research. It's free on Pro, Max, Team, and Enterprise plans through June 22 ; after that it's 10 dollars per million input tokens and 50 per million output . In high-risk areas (cyber, bio, chem) it refuses and falls back to Claude Opus 4.8 — about 95% of Fable sessions run entirely on Fable. This dropped just days after Anthropic publicly warned that AI was getting too dangerous. Why it matters: the strongest Claude is free to try on your existing plan for a two-week window. Run your hardest real task on it now and benchmark it before June 22 — the kind of jump that's worth re-checking your evals against. 2. Apple made its Foundation Models free for small developers At WWDC 2026 , Apple gave developers in the App Store Small Business Program (apps under 2 million first-time downloads ) free access to the next generation of Apple Foundation Models running on Private Cloud Compute — removing inference cost as a barrier. The Foundation Models framework now supports image input . A single Swift API can also call third-party models like Claude and Gemini, server-side. A new Dynamic Profiles system supports multi-agent workflows, and Apple will open-source the framework later this summer. Why it matters: you can ship AI features into an app without an inference bill. Prototype on Apple's free on-device models, and route

The generation side is shipping fast (TileGym, AutoKernel, KernelEvolve). The validation-side surface for “what the kernel actually did at runtime” has not kept pace. TL;DR In the past nine months, three significant releases have landed for auto-generation of CUDA kernels: NVIDIA TileGym , RightNow AutoKernel, and Meta’s KernelEvolve. Each ships training infrastructure for kernel generation. Validation infrastructure (what the generated kernel actually did at runtime, on a real workload, in a production-shaped environment) has not kept the same pace. eBPF traces are the ground-truth layer that closes the gap. What “validation” means at the kernel level Two distinct validation surfaces: Pre-launch: the generated CUDA C compiles, the PTX assembles, the kernel passes a numerical-equivalence test against a reference. Standard compiler / unit-test territory. Generation frameworks ship this themselves. Post-launch: the kernel ran, returned, took N microseconds, used M registers per thread, hit X cache miss rate, and did or did not serialize the rest of the stream behind it. This is the layer that an eBPF trace plus standard CUDA driver counters can answer for any kernel, generated or hand-written. Auto-generation pipelines do not by default close the post-launch loop. They demonstrate “the kernel works in our test setup”. They do not demonstrate “the kernel does not regress p99 latency on production inference traffic”. What an eBPF trace adds to a generated kernel Once a generated kernel is in a real workload, the same trace surface used for any CUDA kernel applies: launch latency from cudaLaunchKernel , sync stalls from cudaStreamSynchronize , host-side overhead from the dispatcher, host scheduling preemption while the GPU is busy. None of those signals are visible to a generation framework that evaluates kernels in isolation. -- post-launch validation: did the new generated kernel regress p99? SELECT kernel_name , COUNT ( * ) AS launches , AVG ( duration_ns ) / 1 e3 AS

Turn AI-app feedback into agent-ready patch context. Discussion | Link

We train a model on handwritten digit classification. 99% accuracy . Then we train the same model on a new task — say, fashion item recognition. We go back and test it on digits. 34% accuracy . It has completely forgotten. Not gradually, not partially — almost entirely. What Just Happened? We trained a CNN on MNIST digits — 99.2% accuracy . After fine‑tuning on Fashion MNIST, it reached 91.1% accuracy . But when re‑evaluated on MNIST, accuracy collapsed to 33.9% . This collapse is catastrophic forgetting : the model’s weights shifted to optimize for the new task, erasing the old solution. Why did training on more data make the model worse at something it already knew? MNIST is handwritten digits (0–9). Fashion MNIST is clothing items like shirts and shoes. Both are 28×28 grayscale images, but the tasks are distinct. Why Does It Happen? The core issue is that the model relies on the same set of weights for both tasks. There is no separation or dedicated memory; every parameter is shared . When training shifts from Task A ( MNIST digits ) to Task B ( Fashion MNIST ), gradient descent simply minimizes the loss on the data it sees at that moment. It has no awareness that Task A ever existed. In the loss landscape, imagine two parabolic bowls: one for Task A and one for Task B. The optimum for Task A lies at θ A ∗ ​ , while Task B's optimum is at θ B ∗ ​ . As training on Task B progresses, the weights θ move towards θ B ∗ ​ . This movement inevitably raises the loss for Task A because its minimum is left behind. The root cause is the shared weight space. Gradient descent is a stateless optimizer; it only follows the current gradient signal. Since the minima for Task A and Task B are far apart, there is no single configuration of θ that satisfies both tasks simultaneously. This is why catastrophic forgetting occurs. Weight space can be visualized as an N-dimensional space, where each axis corresponds to one parameter. Every point in this space represents a full set of wei

When I built whatsmy.fyi , I assumed I'd need a geolocation provider: MaxMind, ipinfo, ip-api, pick your poison. They all mean the same thing: an external dependency, an API key, a quota, added latency, and someone else's server seeing your users' IPs. Then I found out Cloudflare Workers makes the whole category unnecessary. The cf object Every request that hits a Cloudflare Worker carries a request.cf object, populated at the edge before your code even runs. No lookup, no latency, no key. Here's what's inside: { asn : 34984 , // ISP's autonomous system number asOrganization : " Superonline " , // ISP name city : " Istanbul " , region : " Istanbul " , country : " TR " , continent : " AS " , isEUCountry : undefined , // "1" if EU, undefined otherwise latitude : " 41.01380 " , // string, not number! longitude : " 28.94970 " , postalCode : " 34000 " , timezone : " Europe/Istanbul " , colo : " IST " , // which CF datacenter handled this clientTcpRtt : 12 , // user's RTT to the edge, in ms httpProtocol : " HTTP/3 " , tlsVersion : " TLSv1.3 " , tlsCipher : " AEAD-AES128-GCM-SHA256 " } That last group surprised me most: you get the user's HTTP protocol, TLS version, and actual TCP round-trip time for free. Try getting that from a geo API. A complete IP endpoint in ~30 lines export default { async fetch ( request ) { const cf = request . cf ?? {}; const ip = request . headers . get ( " CF-Connecting-IP " ); return Response . json ({ ip , city : cf . city ?? null , country : cf . country ?? null , isp : cf . asOrganization ?? null , asn : cf . asn ?? null , timezone : cf . timezone ?? null , lat : cf . latitude ? parseFloat ( cf . latitude ) : null , lng : cf . longitude ? parseFloat ( cf . longitude ) : null , protocol : cf . httpProtocol ?? null , tls : cf . tlsVersion ?? null , rttMs : cf . clientTcpRtt ?? null , }); }, }; That's the entire backend. No database, no GeoIP file to update monthly, no vendor. The gotchas (learned the hard way) 1. Coordinates are strings. lati

What Is a Claude Code Configuration File? A Claude Code configuration file is a structured file — either CLAUDE.md (a Markdown document) or settings.json (a JSON schema file) — that controls how the Claude Code AI coding assistant behaves within a project or organization. These files define the agent's permissions, memory context, tool access, allowed shell commands, and behavioral guardrails. Without them, Claude Code operates with broad defaults that may not align with your security posture or project conventions. Claude Code reads configuration from multiple locations in a defined hierarchy: a global user-level ~/.claude/settings.json , a project-level .claude/settings.json at the repo root, and one or more CLAUDE.md files that can be nested in subdirectories. The agent merges these at startup, with project-level settings taking precedence over global ones. Understanding that hierarchy isn't optional — it's the foundation of any serious deployment. Why Claude Code Configuration Files Matter in 2026 Claude Code has moved from a tool used by individual engineers to something teams are deploying org-wide, running in CI/CD pipelines, and integrating with production infrastructure. That shift changes the risk profile completely. A misconfigured agent with shell access and no guardrails isn't a productivity tool anymore — it's a liability. Anthropic's own documentation on Claude Code security acknowledges that the agent can execute terminal commands, read and write files, and make network requests. By default, many of these capabilities require per-operation approval, but configuration files can silently expand those permissions across an entire organization if applied at the global or enterprise policy layer. The CISA and NSA joint guidance on AI-assisted development tools (published in late 2024) specifically flagged AI coding assistants as a new attack surface for supply chain compromise — the concern being that an agent with write access to source files and no beha

OAuth for Remote MCP Servers How each AI assistant signs in to a remote MCP (Model Context Protocol) server, and why the flow differs by client and by where it runs. Overview The protocol throughout is standard OAuth 2.1 — an open, widely implemented authorization standard. The human sign-in runs through oauth2-proxy , one of the most widely deployed open-source auth proxies; the only deployment-specific piece is a thin, spec-conforming authorization server (the /oauth endpoints) that hands MCP clients their tokens. Every client ends up the same way — a person signs in against Google (restricted to your organization's domain), and the client holds a short-lived bearer token it presents on each /mcp call. Two things differ between assistants: where the client runs (a machine on the VPN — private — vs. the vendor's cloud — public ), which decides the host it reaches; and what kind of OAuth client it is — a public client proving itself with PKCE (Proof Key for Code Exchange, which lets a client with no secret prove the token request comes from the same client that started the flow), or a confidential client proving itself with a secret. The participants oauth2-proxy — the public-facing reverse proxy. It authenticates the human against Google (the sign-in restricted to your organization's domain) and forwards the verified identity to the app behind it. Only oauth2-proxy faces the internet. It is a mature, heavily-deployed open-source project — the standard way to put Google/OIDC (OpenID Connect) single sign-on in front of a service, widely used in Kubernetes deployments — so the most security-sensitive leg of the flow (the OAuth exchange with the identity provider) runs on battle-tested code. The MCP server — the app on a loopback port behind the proxy. It plays two roles: the OAuth authorization server ( /oauth/authorize , /oauth/token , /oauth/register , .well-known discovery) and the /mcp tool endpoint. It mints codes and tokens, and validates a token on every /mcp c

Users under 16 years old will get a separate profile to show Stories and Spotlight posts to friends that they follow back.

Electricity demand from AI data centers is pushing everyone — including automakers like GM and Ford — into the energy storage business.

I spent a weekend wiring Google's Gemini and Veo APIs into a single app just to feel where the edges of multimodal AI actually are. It turned into a small studio I now use daily, and along the way I learned more about these models from plumbing them than from any paper. Here's the honest technical debrief. Three pipelines, three completely different problems I wanted one prompt box that could do video, image editing, and document Q&A. Naively I assumed they'd share most of the stack. They don't. 1. Image-to-video: the enemy is time, not pixels Generating one good frame is solved. Video is about temporal coherence — frame 13 must agree with frame 12 or you get flicker and identity drift. Modern video models treat the clip as one object in space and time (latent diffusion over a width x height x time volume, with spatiotemporal attention) rather than 120 independent images. Conditioning on a reference image as the first frame is what makes image-to-video feel controlled: you've handed the model a strong anchor and asked it to extrapolate motion, not invent a world. The surprise: native audio sync (Veo 3.1 generating clip + soundtrack jointly) does more for perceived realism than another notch of resolution. A door slam landing on the exact frame the door shuts is uncanny in a good way. 2. Instruction-based image editing: preservation is the hard part Generating is unconstrained; editing must change one thing and preserve everything else. Condition the diffusion model on both the instruction and the source image's latents, cross-attend the instruction to steer only the referenced region, and bias hard toward preserving unedited latents. Push that preservation too soft and the subject's face quietly morphs across edits — the classic 'character consistency' failure that makes or breaks storytelling use-cases. 3. PDF chat: it's retrieval, not a long context The naive 'paste the whole PDF' approach dies on long files (models get lost in the middle ) and costs you the full

Hello everyone! Following up on my previous post , Day 1 of my Modern Data Stack migration was an absolute rollercoaster of refactoring and deep data auditing. I’m moving our legacy system (spreadsheets and Qlik) into a robust pipeline using Python, ClickHouse, and dbt . Here is what went down over the last 24 hours. 1. From Messy Scripts to a Single, Parameterized Extraction Engine 🛠️ In the legacy setup, each company had its own folder, its own .env file, and its own duplicated Python extraction script. It was a maintenance nightmare. Yesterday, I completely refactored this structure: Centralized Configuration: Merged all separate environments into a single, global .env file at the root level, mapping all 8+ companies and their branches. Eliminated Code Duplication (DRY): Instead of having identical extraction logic copied across folders, I built a single, unified codebase. Now, we have one universal script for Sales, one for Stock, one for Orders, etc. The behavior changes dynamically based on the company argument we pass to the CLI (e.g., python -m extract.run extract --source company1 ). To speed up this refactoring, I used Claude to generate the initial application skeleton. Since the AI already had the context of our legacy extraction logic, translating it into this new clean architecture was incredibly smooth. 2. Highs and Lows: The Data Parity Challenge With the pipeline modernized, I ran the pilot ingestion for Company #1 . To minimize friction for our downstream BI consumers, I kept the ClickHouse Bronze tables structured 1:1 with the legacy CSV schemas. The Good News: The data ingestion into the Bronze layer worked flawlessly. Moving up to the Silver layer (where we do data cleaning and domain-specific transformations), everything validated beautifully. Row counts matched perfectly. The "Fun" Part (The $2 Million Gap): When I materialized the Gold layer (our consolidated group business models), I hit a massive wall. The new pipeline reported $2 million U

Waymo made a virtual driver called ReD to help its robotaxis avoid accidents.