Are companies actually moving away from LeetCode?
Yes, and in some places it raises the stakes rather than lowering them.
From our own interview data, yes. DSA is still a convenient first filter for big companies, but as memorization and cheating tools get better the companies are pushing back with more creative and more original problems, and over ninety percent of startups and quant shops now run at least one implementation or design-style round, because those map far closer to the actual work. So DSA is not dead, it simply stopped being enough on its own, and the people who can also reason through fresh, implementation-heavy problems are the ones who now stand out from everyone else clearing the same filter.
Why this matters more in India
Where you are changes the stakes entirely, and this is the part the influencer crowd has backwards. In the US, DSA is one signal among many, so if you are weak at it there are other ways to be seen: you can build a real product, ship open source, do well at a hackathon, do interesting AI work, or get a referral, and the interview is not the only door into the building. In India, for most people, it genuinely is the only door, because the pipeline runs through campus placements and placements run on DSA, so you either clear the bar or you never get in front of anyone at all. That makes the quality of your DSA practice matter more rather than less, since it is the single thing your entire outcome hangs on. The rote sheets, the coaching-factory advice, and the N patterns solve everything content therefore do the most damage exactly where people can least afford it, because the market that most needs good DSA habits is the one being sold the worst ones. Getting the early stage right, real reasoning instead of pattern matching, is worth the most precisely in a system where DSA is the only thing that gets you seen.