How to keep AI spend flat while token usage grows exponentially: Not with friction and spend alerts. With better defaults, routing, and caching. Better Defaults (not Usage Caps) – Engineers can choose any model they want, but defaults matter. We’re experimenting with defaulting to open weight models like GLM 5.2 and Kimi 2.7 through our LLM gateway, while still encouraging engineers to choose the right model for the task. 91% of our employees were never hitting their usage caps, so instead of lowering caps and driving up alerts, we're moving to cheaper defaults. Note that code reviews use a diversity of models, so they can check each other's work. Better Routing – In our custom harnesses, we preprocess prompts and route to the best model for the job, considering cache hits and model pricing. For instance, you may want a frontier model for planning, but not for execution where they can be overkill. Ultimately, humans shouldn't be choosing models - AI can automate this task. Better Caching – Cache misses are the easiest way to drive your cost up. All of our requests are cache aware, so we’re reusing a warm cache wherever possible. For example, our cache hit rate went from 5% → 60% in LibreChat once properly implemented. Keep Context Lean – Start fresh sessions when switching tasks. Scope file context narrowly. Disconnect unused tools. Don't just compact. The goal isn't fewer tokens used, it's fewer tokens wasted. Better Visibility – Our engineers can use as many tokens as they want, from whatever model they want, but we’ve made usage visible – and the more you spend on AI, the more impact we expect. The goal isn't to suppress usage. It's to build the infrastructure that makes exponential growth sustainable. Putting this into practice has cut our AI spend nearly in half, while our token usage continues to grow.

yes claude dropped tag, but we had this months ago @coinbase - shout out @alrodi & @kylecesmat's team & more😉 My team at Advisor now has our own branded Slack agent. Full context on all docs, tickets, codebases, and observability and triage ability. Our new employee.

Big week for Coinbase! Like I said on stage, thanks to all the Coinbase employees (and their thousands of AI agents) whose hard work and dedication made all of our announcements possible. The everything exchange now includes pre-IPO perps, stock options, and tokenized stocks soon. We also redesigned Coinbase Advanced, and started to combine our global liquidity (between US and international users, and also Coinbase and Deribit users) @CoinbaseDev is bringing the benefits of stablecoin payments to businesses everywhere, with fully custodial accounts using our compliance stack, and launched an awesome new dashboard for all dev tools. For @base, we announced private transactions and Base App on web. And finally, Coinbase is also becoming the financial account for AI - give your agent a wallet, get AI-powered financial advice, and connect your Coinbase account to your favorite LLM.

1. as a mental model it is more correct to think of fable+ class models as english -> code interpreters - converts your idea into code into "correct" code regardless of problem complexity and output complexity (diff size). Fable 5 will be the worst of this new class of models 2. diff size/complexity is to be managed purely for review: small diffs - in high risk areas of code (auth/identity/data access/network access/money movement) large diffs for code that can be empirically verified (frontend/backend plumbing/code without network or db access/performance code that can be empirically verified) 3. time it takes to ship software is completely disconnected from time to produce the PR - how long the work takes depends fully on ability to review/merge code while managing risk at scale 4. solving the bottlenecks for above matter enormously- linters/testing/CI/shadow mode verification/empirical verification 5. agency matters enormously- what are the biggest bottlenecks to speeding up the loop and eliminating them? what are the problems that need solving and when do they need solving? what does it take to the solution to all of them today? 6. deep understanding of the full stack matters enormously- what problems are worth pursuing? is there a higher level of problem abstraction to address first? should I give it the sub-sub task, the sub task, or the task itself. what are the major risks with this PR (order of importance: security holes/correctness holes/performance holes). is there a higher speed way of producing data that allows me to merge this? should this be run in shadow or in a sandbox or a flag. understanding every line of logic may not be needed but understanding and managing risk matters enormously. 7. the cost of complexity itself is changing. it might be now worth "maintaining" 50% more code to get a 5% performance win. getting the right abstractions matter less because larger refactors are less tedious. code quality nits become huge drag. very likely, a much smarter model will be maintaining your code so worth taking on more technical debt now. taking the time to hand architect and rebuild systems comes with an enormous cost of velocity 8. if it quacks like a duck and walks like a duck, it's a duck. For low risk cases, it might be more sane to treat code chunks (services / functions) as a black box, like we do for neural networks: do full empirical verification only: has code produced correct outputs for the last 10,100,1000,10k inputs ? can we quarantine this large piece of code - no outbound access to network / database ? what happens when this code is wrong? do we get hacked/or crash(memory/cpu)/is an inconvenience? is it internal facing or external? what can we do to address these risks? 9. eventually, logical verification (line by line review) will come at an enormous cost- save it for where it matters and build systems that are tolerant to empirical verification. is there a decorator that prevents db / network access? correctness bugs are significantly easier to rectify than access bugs 10. what are the rails that allow for even faster iteration? code permissions can be opt in - db writes, db reads, network egress (to where?), PII access. how long does it take to get shadow mode data? how many PRs can be tested? What are the categories of diffs

Travel the world, and earn Bitcoin for doing it. We’re launching a travel portal for Coinbase One Card holders. Get 5% back in Bitcoin with each travel purchase through the new travel portal.

We started as a place to buy Bitcoin, now we power your entire financial life. Here’s everything we announced today ↓ → Tokenized Stocks → Pre-IPO Perps → Stock options → Crypto options → Perpetual-style equity indices → Crypto derivatives, back in America → Time-based prediction markets + Combos → Launches: access millions of tokens, the moment they go onchain → Coinbase Advisor: an AI investment advisor, built into the app → Coinbase for Agents: connect any AI agent to your account → Base MCP + x402: give your agents their own wallet → The new Coinbase Advanced - fully modular → One unified global liquidity pool → Coinbase One Card, now more accessible secured by USDC → 5% Bitcoin back on travel → Crypto-backed mortgages → Borrow against your staked ETH & SOL with new liquidation protection → Transfer Protections → New Coinbase Developer Platform → Full stack Coinbase Payment solution → B20: a Base native token standard for any asset → Base App on web + multichain support + all assets → Private transactions for enterprises on Base See you next time.

SITUATION DETECTED: Andrej @Karpathy has joined Anthropic.