Every time I read stuff like this I honestly wonder if the author is using the same tools I am.
I can have Claude Code bang out everything from boilerplate to a working prototype to a complex algorithm embedded in a very complex and confusing code base. It’s not correct 100% of the time but it’s pretty damn close. And often times it comes up with algorithms I would have never thought of initially.
These things are at least a 10x multiple of my time.
What tools / experiments out there exist to exercise these cheaper models to output more tokens / use more CoT tokens to achieve the quality of more expensive models?
eg, Gemini 2.5 flash / pro ratio is 1 1/3 for input, 1/8 for output... Surely there's a way to ask Flash to critique it's work more thoroughly to get to Pro level performance and still save money?
This looks very interesting, however it seems to me like the critical piece of this technique is missing from the post: the implementations of getFileContext() and shouldStartNewGroup().
This is awesome. Love this idea. What a great way to make history more alive. I've only spent ~30s with it so far but I hope to find ways to contribute to it (content and code/different visualizations)
Do you have any open source examples (either your own or others) of simple Ruby + Sequel that are particularly elegant + productive that you could point to? Would love to see this.
I agree Ruby is the most elegant and enjoyable language to develop in.
Wow. So if this is universally true - that all non-macs are garbage - why is it that review sites like Notebookcheck rate some non-macs machines close to (albeit not matching) a Macbook?
Also with their packaging tech (EMiB and Foveros) they can mix and match chips from different fabs - ie, combine P-cores from TSMC fabs with E-cores from their own (possibly older) fabs. Alder Lake is the first real stab at the "chiplet" architecture, and it seems to be a very good outing. Future iterations will only improve, and at Intel's purchase scale economics they'll be able to manufacture chiplet designs that match market needs for the highest margin or pricing pressure to squeeze competitors, if needed.
This is a "heads I win, tails I win" strategy. Give the fab business a chance to be independent and justify its massive capex and lower margin profile. Increase utilization of nodes to squeeze as much cashflow out of them as possible. If it continues to be a worthwhile cashflow endeavor, fantastic. If not, split the company and fetch a high price for a sophisticated and capable (if not #1) semi fab business.
Let the higher margin chip design business utilize whatever manufacturing process is most effective for its latest and greatest chip designs. Chip design and development won't be hamstrung by setbacks in manufacturing. Intel's scale will ensure most optimal pricing and deal terms with TSMC (if INTC doesn't use its own fab).
Along the way, there is additional upside optionality from government subsidies and support for a critical piece of infrastructure in the 21st century. Domestic fabs will be (if they aren't already) as high priority and critical to US national defense as the military hardware contractors (Boeing, Lockheed Martin, etc). That also ensures a floor to the price of the fab business should Gelsinger decide it should be sold.
I want this to succeed, so I want the company to succeed. On that note I think you guys should change up your pricing.
Seems like there's too big a gap b/w the free forever (individual) and $20/mo/user for team.
I'd love to pay for this as an individual at a smaller amount - like $10/mo - for a few extra features. Or maybe reduce the functionality of free forever.