Claude writes about 90% of Claude Code's code.
From June to September, merged PRs more than doubled while the share reverted fell by more than 40%.
We wrote about how we changed code review to get there, and how to adopt it:
https://t.co/9sZe64g3mb
the focus on whether you read the code or not is the wrong thing to look at
if you have software you're responsible for, you should be able to answer questions from memory about how it works
the expectations for how well you can do this should not be any different now
👋 New blog post:
"How coding agents read files – and how to write for them"
In it I explore how code optimized for agent-discovery may lead to major token savings on search + retrieval, even in code generated from frontier models like GPT-5.6 Sol
Plus other findings ⬇️
"I think it's important for people to understand how code works."
Geoffrey Litt's Design Eng track keynote is live now:
https://t.co/AFQnJA4mqK
Thank you @NotionHQ for supporting his incredible work and thank you Geoff for kicking off WF26 for our Design Engineers!
Hot take: I think it's still important to understand the code that our agents write!
In this mega thread (based on my AIE talk today), I will explain why that's the case, and show some ideas for how to efficiently understand code. Alright, let's dive in. 1/
I recommend using GPT-5.6 Sol Medium as your daily driver, and for really hard problems to switch to Extra High.
If you want the absolute best and are not afraid to burn usage fast, then Ultra is a beast.
We benchmarked coding agents on real engineering tasks across Databricks' multi-million line codebase to better understand the relationship between cost and performance.
The benchmark surfaced several key findings. Top-tier performance now comes from a mix of proprietary and open models, token pricing is a poor proxy for real-world cost, and harness choice can dramatically impact both cost and quality.
Read how we built the benchmark and what we learned → https://t.co/W0Dxhvj1zM
I had some vibes that Opus 4.8 was performing worse than older ones for some of uses that are off distribution and now I have the receipts. Latest Opus/Sonnet are causing tool invocation failures on Pi's edit tool when older ones did not! I wrote about it. https://t.co/96Qd8zfdVc
Two-pizza culture was never about feeding hungry engineers. It was about ownership, speed, and not being bogged down by bureaucracy. Working backwards from the customer and putting pen to paper have always been our way of forcing clarity of thought. But the industry is changing, and it’s time to reconsider how we bring products to live. Read more: https://t.co/TIfpLLWRNQ
@wasabigeek In terms of tools, my team at @StashAway_SG developed a financial planning feature to visualize progress towards financial freedom. We also implemented CPF handling in our projections.
https://t.co/tdPoHlbkPh