In partnership with
100+ Claude Code hacks to ship code 10X faster
Top engineers at Anthropic and OpenAI say AI now writes 100% of their code.
If you're not using AI, you're spending 40 hours doing what they do in 4.
These 100+ Claude Code hacks fix that and help you ship 10x faster.
Sign up for The Code and get:
100+ Claude Code hacks used by top engineers — free
The Code newsletter — learn the latest AI tools, tips, and skills to code faster with AI in 5 minutes a day
TODAY IN AI
OpenAI announced that its internal model has resolved more than 100 long-standing open problems across most of mathematics, and formed an independent advisory group of nine mathematicians, including Timothy Gowers and Edward Witten, to assess results and advise on publishing them. The group is unpaid and can speak freely, but TechCrunch notes it has no say over how fast OpenAI moves. None of the 100 has been independently verified yet, and 25 Fields Medalists signed a letter this month objecting to labs racing on famous problems.
Amazon blocked Meta's Muse agent from shopping on Amazon.com, saying it never asked permission, does not identify itself when it browses, and stores customer credentials. Meta says the agent never sees passwords. Amazon made more than $68 billion from ads last year, nearly all from people scrolling past sponsored products that an agent would skip.
Shopify went the other way and connected Muse to Shop Pay, so the agent can find and buy products across Shopify stores in one tap, inside Muse or WhatsApp. If you run a Shopify store, a Muse user can now reach your checkout; it is worth deciding whether you want that before the holiday season.
Google unveiled the Googlebook, a laptop line from HP, Dell, Lenovo, Acer and ASUS starting at $899 and pitched as the first laptop designed for Gemini. Wiggle the cursor and Gemini appears to act on whatever you select; a feature called Rambler turns rambling dictation into clean, formatted text. Preorders are open with no ship date yet.
SpaceXAI released Grok 4.7 on a new, larger base model at the same price as 4.6, with a fast variant at double the speed for double the cost. It is live in Cursor, Grok Build and the API.
Apple is paying out a $250 million settlement to US buyers of the iPhone 15 Pro and iPhone 16 models who bought between June 2024 and March 2025 expecting the delayed Siri features. Claims are $25 per device, possibly up to $95, and the deadline is December 21.
TUTORIAL
How to create a project intelligence database with Notion AI
This takes about twenty minutes to set up and then maintains itself as long as your team keeps dropping notes into the project pages.
Open Notion, create a new page, choose Table, then select Build with AI.
Ask Notion AI to create a project database with fields like Project, Owner, Status, Due Date, Priority, Summary, Risks and Keywords. This prompt works:
Create a project tracker for my team with Project, Owner, Status, Due Date, Priority, Summary, Risks, and Keywords. I want AI to help summarize each project and surface potential risks from the information stored in each project page.
Add each project as a database entry and put the relevant briefs, notes, decisions and updates inside its page. The AI fields read from whatever is in the page, so the more you put there the better the output.
For Summary, open the property settings and choose AI Autofill, then Basic, then Summary.
Use AI Autofill for Keywords and any other structured field you want Notion to extract on its own.
For Risks and Priority, choose AI Autofill, then Custom Agent, and give it an instruction like:
Identify the biggest current risk and classify priority as High, Medium, or Low.
Set each AI property to update manually, when the page changes, or on a schedule. Weekly on a Monday morning works well for a team tracker.
The risk and priority fields are only as honest as the notes underneath them; if nobody writes down that a project is slipping, the database will not know either.
TOOLS WORTH TRYING
4 tools catching our attention this week.
Clarify: Sponsored. An AI-first CRM that folds prospecting, follow-ups and note-taking agents into one system, pitched as a replacement for four or more separate sales tools. Best suited to small sales teams; larger ones should check the migration path from their current CRM first.
Hoogly: Turns employee voice notes and feedback into tracked actions with AI. Useful for managers who collect a lot of verbal input; it depends on people actually recording their thoughts.
Sai: An AI secretary that uses its own computer to do your screen work, from filling forms to moving data between apps. Powerful but early, so start it on tasks you can check afterward.
Streva: Instant translation anywhere you type, across apps. Good for quick messages with overseas suppliers or clients; for contracts, still use a human.