The Archive

Uber Caps Usage of AI Tools Like Claude Code to Manage Costs I wrote the other day about Uber blowing its 2026 AI budget in four months, and how that wasn't particularly surprising given they would have set that budget in 2025, before anyone could have predicted how popular token-burning coding agents were about to become. Natalie Lung for Bloomberg: The rideshare giant is limiting all employees to $1,500 in monthly token spending per AI coding tool, an Uber spokesperson said in response to a Bloomberg News inquiry. That means spending on one tool doesn’t have a bearing on the budget for anot...

How we contain Claude across products A complaint I often have about sandboxing products is that they are rarely thoroughly documented , and in the absence of detailed documentation it's hard to know how much I can trust them. Anthropic just published a fantastic overview of how their various sandbox techniques work across Claude.ai , Claude Code, and Cowork. We constrain where and how an agent can act with process sandboxes, VMs, filesystem boundaries, and egress controls. The goal is to set a hard boundary on what an agent can reach. For example, if credentials never enter the sandbox, they...

Research: Running Python ASGI apps in the browser via Pyodide + a service worker Datasette Lite is my version of Datasette that runs entirely in the browser using Pyodide in WebAssembly. When I first built it four years ago I used Web Workers and code that intercepts navigation operations and fetches the generated HTML by running the Python app. This worked, but had the disadvantage that any JavaScript in <script> tags would not be executed - breaking some Datasette functionality and a whole lot of Datasette plugins. This morning I set Claude Opus 4.8 the task (in Claude Code for web) of figu...

Are AI agents tools, co-authors, or researchers? We present a quantified case study ($N=1$): a physicist supervising an AI coding agent (Claude Code, Sonnet and Opus models) over 12 work days and 57 sessions to build CLAX-PT, a differentiable one-loop perturbation theory module in JAX. We documented and classified 15 supervision events by intervention level. The agent resolved ten autonomously by iterating against oracle tests. Two more by the physicist's domain knowledge. The three it could not -- all evaded oracle detection -- share a common property: the agent treated symptom reduction as ...

Today we're introducing dynamic workflows in Claude Code. Claude now writes its own orchestration scripts, fans work out across tens to hundreds of parallel subagents in a single session, and verifies its own results before anything reaches you. Work you'd normally plan in quarters can finish in days. Built for the tasks a single pass can't handle: codebase-wide bug hunts, security and optimization audits, large migrations and language ports, and high-stakes work where you want adversarial agents trying to break the answer before you see it. Progress is checkpointed, so long runs survive int...

We’re upgrading Claude Opus to a new version: Claude Opus 4.8. It builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors. Available today for the same price. In Claude Code, you can hand off a feature, a migration, or a bug sweep and let it follow the work through while you focus on what’s next. Also launching today: * Fast mode for Opus 4.8 (research preview). Same model at roughly 2.5x the speed, now three times cheaper than before. * Dynamic workflows in Claude Code (research preview). Claude ...

The reported Uber story is so on brand it almost reads like satire. Incredibly useful tool, slightly magical workflow, then finance walks in with a flamethrower in April. If they really finished the year's claude code budget by month four, that does not mean claude code is bad. It means the usage pattern changed faster than procurement math did. Claude is good enough at coding that people stopped treating it like autocomplete and started treating it like a coworker that never sleeps. That is exactly where the cost curve gets weird. A dev asks for a refactor. Claude reads context, plans, edi...

At work we've been prompted about running Claude Code overnight. The suggestion came in form of a document that loosely outlined how this could be done... use git worktrees, make tight specs, no commit to main, static code analysis and lining etc. Very high level. Had a bit of sales pitch smell to it, but has enough content to peak my interest in spite of it. I looked at reddit to verify if this is even an idea that could be taken seriously. I could only find a couple of reddit posts with little actual information and usually from about 4-6 months ago so not much credibility for today. I'd ...

I've been noticing this lately. I use Opus 4.7 with Claude Code, and I've been using Claude Code for a long time. Lately, I've been noticing some strange behaviour from Opus. Things like; \- Stopping for no reason and asking "should we stop here?" in the middle of a task \- Asking multi-choice questions with a "pause here, I'll continue later" included in the options randomly for no reason \- During a requirement-gathering questionnaire, asking me "why do you need this" and "what would you do if this feature was not implemented?" (it asked me this today and I was really surprised by thi...

no more separate usage limits.

Every time I watch someone use Claude Code on a real codebase, the same thing happens. It rewrites a module that three other modules depend on without any awareness of coupling. It just reads the file, makes changes, moves on It reads files one at a time without any map. Doesn't know which files are coupled. Doesn't know who owns what. Doesn't know why that weird pattern in the auth module exists on purpose. I've been building an open source MCP layer to fix this called repowise. Self-hosted, pip install, AGPL-3.0. Five context layers that sit between your codebase and the model: Graph - ...

Quick story: I kept losing context every time I had to \`git stash\` and switch branches to test something an agent had suggested. Then I actually read up on \`git worktree\` and it solved the whole problem. What my setup looks like now: \- Main worktree: where I review and commit \- 2-3 extra worktrees, each with its own Claude Code session running on its own branch \- When one agent finishes a task, I \`cd\` in, review the diff, merge, move on \- No stashing, no context switching, no "wait what was I doing" Full writeup in the article on Medium (https://medium.com/@buildwithpulkit/git...

I'm going to describe a person this post is for, if this is you, I think I can be of some assistance: * you are new to coding * you are blown away by how it unlocks this magical ability that was previously inaccessible without years of training and effort * you've daydreamed of business and app ideas but never knew where to start before or how to build them * you've been vibe coding non-stop and burning through tokens * you're unsure about what's secure, how to structure the systems, and how systems are supposed to interact with each other. So, essentially the plumbing separate from the code...

My company started tracking Claude Code usage - tokens and spend, that kind of thing. Now my manager wants me to stack-rank my engineers on "AI performance" using those numbers. I'm not comfortable with it (but I don't have a choice either). Token usage feels like exactly the wrong proxy - my strongest engineer uses Claude surgically while someone burning 10x the tokens isn't 10x more productive (often the opposite). Ranking on this just teaches people to game the metric. So, for folks here who use Claude daily and/or lead teams: * Has your company started measuring "AI performance"? How a...

https://x.com/i/status/2059298565093196012

This is especially prevalent in Claude Desktop and less obvious in Claude Code. * Often leads long assets with a BLUF (bottom line up front), specifically military-adjacent terminology. * Has slipped into language where it refers to civilians several times, as though it means someone other than us * References to things like deep-dive, vision-intent framing, force multipliers, and reframes are more subtle because they do cross over but their prevalence is unmistakable. This isn't a complaint, it's more a note that, "hey guys, we see your training material in our outputs" and its interestin...

Microsoft, has started canceling Claude Code licenses, per the Verge

Uber president Andrew Macdonald (pictured) says its “hard to draw a line” between AI spending and deliverable features. | Photo: Zed Jameson/Bloomberg via Getty Images After reportedly exhausting its annual AI budget just four months into 2026, Uber is now questioning whether it's actually seeing meaningful returns on its investments. In an interview with Rapid Response, Uber president and chief operating officer Andrew Macdonald said the company isn't seeing a connection between rising token consumption for Claude Code and more useful features being delivered to consumers. "That link is not ...

Hello, I'm on Windows having setup both Claude Code App and Terminal, but I find the App simply more convenient to use. I have had several people pushing me to use the Terminal saying "the App is low" and "Terminal is so much better" ... but when I inquired none of those people could actually name a single thing that the App would be missing (everything they mentioned the App has as well) or a single concrete reason why I should switch to Terminal beside vague phrases So is the terminal substantially better than the App in something, are there reasons to switch besides being used to it and p...

I noticed that Claude Code was giving me a lot of unsolicited caveats with phrasing like "honest caveat" or "genuine caveat" when this kind of hedging was absolutely unnecessary. I figured other people might be seeing the same thing so my instinct was to use Google Ngram but the cutoff year of 2022 meant that I had to use a different method. So I used Google search with quotes around the phrase "honest caveat" and set the time bound to different time intervals and compared the number of search results as a proxy for how usage has changed over time in indexed pages. As it turns out, while del...

Reddit user reports erratic behavior in Claude's code generation; lacks technical details or reproduction steps.

Lawyer documents local V100 cluster setup (12×32GB SXM2) for in-house legal document drafting using Claude Code.

Found a really clean Claude Code use case that's not coding-coding. Remotion (React for video) means motion graphics are JSX components. So I describe what I want in plain English, Claude Code writes the component, I render. Lower thirds, intros, overlays, all reusable across videos. Iterations are seconds instead of the typical "drag clips around in CapCut for an hour" loop. Visual style is finally consistent across my channel because the components are shared. 13 min walkthrough on my channel, link in comments to avoid spam vibes.

Developer built iOS/watchOS app and landing page using Claude Code, reached 1,500+ users by solving friction in law enforcement location-sharing workflows.

For anyone using Claude Code, codeMeter is a small WiFi desk display that keeps your usage visible while you work. It shows your 5 hour usage, weekly usage, reset countdowns, and color warnings as you get closer to your limits. No laptop app or browser tab needed once it is set up. Just plug it in, connect to WiFi, and keep building. If anyone is interested in building one , reach out, I am happy to share the source for free. Finished models are for sale at [Encinitas3D.com](https://encinitas3d.com/product/codemeter-a-desk-display-for-your-claude-code-usage/?utm_campaign=reddit-organic)

https://preview.redd.it/b9kl1t77743h1.png?width=955&amp;format=png&amp;auto=webp&amp;s=6ae0f4ac116e08ce619765b6674144961058a689 https://preview.redd.it/ls5oh1be743h1.png?width=416&amp;format=png&amp;auto=webp&amp;s=0e2899df9ef0d704cb5da08b1916f6009ac18b37 Yes, I am absolutely certain that this ban is for no reason. I mainly use Claude Code; I either ask technical questions or simply complete my daily coding tasks, and that is all. The only reason I can think of is that I’ve recently been working on a complex project that requires running up over $100 worth of tokens every single day—so per...

Claude's prompt caching costs 12.5× more on cache miss vs. hit; common mid-session actions trigger invalidation and inflate token bills.

The model became visible for a short amount of time on Claude; besides that, new strings mentioning Mythos have been added. \&gt; Access to the Claude Mythos model in Claude Code and Claude Security. It still doesn't mean the general public will have access to this exact model, according to Anthropic's earlier communication. source : testingcatalog https://preview.redd.it/tb7riwqs8z2h1.png?width=900&amp;format=png&amp;auto=webp&amp;s=743f7570a7a5d8bc662f49ef24060f5e9cde258b