- you create large number of working small apps .

- you create a spec from these apps .

- create a huge app .

- make a Dsl to make extensible .

- extend the Dsl to fit what you need in the future .

- optimize the Dsl remove obvious N+1 stuff.

The hard part is throwing away the code in each step . Both managment and devs cant stomach the reality that the code is useless at each stage prior to dsl. They cant molt and discard the shell and hence the project dies.

comming back to research one of the frontier model's deepseek was able to come close to SOTA with a relatively small budget because of one of their mixture of experts approach.

Please learn

Some emphasis should be made on the total percentage of H1B applicants 75% are Indian and at-least ~90% of those are in the TECH working for CTS, TCS both are consulting firms They dont sell any products or services (and dont have offices in SF). They survive by giving the lowest bid for a service . And when the minimum wage is the lower bound why bother paying any higher (60-75% discount), This was the case prior to trump. After Trump H1Bs became scarce and to justify it you needed "highly skilled" employees and they made the paygap more justifiable to 45-50%.

So i still hold my original statement H1B is for tech what H2A is for agriculture.

H1B is to tech what H2A is for agriculture . Yall want to Drive the combine ( Managment, Software development, VC ) but dont want to do the dirty work ( QA, Devops, IT admin )

-> Getting an assignment in the US is viewed as a reward for employees who work in the company for 5+ years.

-> H1B visa employess are given a 50% minimum discount to market rate. cause lol what are you gonna do . Quit and go back in the Queue for H1B.

-> nepotism / favoritism self explanatory.

-> Are more willing to work 40+ hours a week. Less likely to take vacations.

These are the observations of myself and members of my family who are from India.

reasoning by pattern matching is quicker , when you fail you search for new patterns this is useful when the outcome is of low impact.

both are useful tools from my experience.

carbon credit: EU law fines ICE car makers for making cars. they inturn purchase carbon credits to have the right to keep polluting. further more tesla made 53 B$ in rev for 2021. 1.5B is not nothing but not > 5% of revenue for the year. ie bulk of it came from actually making and selling cars.

tax credit: the US federal govt had a tax credit program which was lobbied by GM in the 2010's that gave car buyers 7.5K as a tax write off . this was only eligible for less than 100k cars sold (tesla sold 405,300 in q4 2022) and was only useful if you made enough income where a tax writeoff made sense. 2023 new laws are in place with more rules which makes them almost useless (or only useful for those who lobbied for it).