The docs say that CLAUDE_CODE_EFFORT_LEVEL controls adaptive reasoning intensity, and CLAUDE_CODE_DISABLE_ADAPTIVE_THINKING bypasses that entirely in favor of a fixed budget via MAX_THINKING_TOKENS. So setting both is contradictory. If true, disabling adaptive thinking would override what effort level is trying to do.
Hi there! That's a really good question. The short answer is: It depends on your desired business logic, your product, and your customers.
>Do you pre-bill them sans shipping address and then hope that they input it later?
In this DIY checkout example, your customers are charged at the time of checkout. Then you could follow up with them en masse to fill in their shipping address before you're ready to ship. This is a self-service feature https://dashboard.trycelery.com/status and the credential parameters can be pre-formed so convenient links can be dynamically created for each customer to take them right to the correct spot.
Once a customer has paid for something, the conversion of them coming back to fill in shipping address to receive it is very high.
>how many backers do you lose by not obtaining actual shipping details upfront?
Bravo. Not many people think of the flow so holistically, even though they should. On one hand, reducing inputs and simplifying checkout can give conversion rates a lift. On the other hand, unfamiliar UX can trip some customers up too. Also it can end up driving more support questions for sellers. eg, "When do I give you my shipping address? How do I input it?"
Overall, we're convinced so far that the shorter checkout is a net gain for conversion.
2% is something we re-evaluate often here at Celery in the continued effort to price fairly. The Celery fees are only charged when the orders are charged so for delayed pre-orders a merchant could be set up in 90 seconds and use our software to capture tons of demand, interact with customers, manage orders, receive guidance and support from our team - all without paying a cent up front.
You're absolutely right in that the 2% makes a lot of sense for the convenience and for the stage a lot of hardware startups begin. We do also offer tiered breaks on the fee at different sales volumes right now and are overall still trying some things on pricing. Really appreciate the feedback.
We've had the chance to work with so many awesome teams like Pebble, Timex, Boosted Boards, and a great group of the biggest projects from Kickstarter. It's hard to list them all but here's a sampling: https://www.trycelery.com/customers
As for custom and DIY checkouts, August.com is one of our coolest examples of a checkout built on our Celery API and there are some exciting ones in the works right now. Stay tuned!
>What sets you apart from the other instant and "easy to set up" checkouts?
In fact, the default product at Celery is exactly one of these "easy to set up" copy-pasted checkouts with just pasting in 2 lines of code to get a checkout. This DIY checkout offers something that none of the copy-paste checkouts, including our own, offers: Total customization over every aspect of the checkout interface and flow.
Not only could you change all the style and position of the elements, you can even insert steps afterwards, like your own survey, for example. You could even change the language to Klingon if you like. We're excited to see what people can come up with.
Branding is totally open to your own styles. Colors, logos, assets, transitions and animations - you name it.
Amazon Payments is a feature that we'll be keeping a keen eye on and prioritize it based on demand and product fit.
Is the Uber pricing working? Sorry if I missed something but it seems stuck at $7–$9 no matter what start and end destination I use in San Francisco. Screenshot http://cl.ly/image/312i1H3g3b2L
https://code.claude.com/docs/en/env-vars