The block of text is the 'Likely to outsource' column. I use Perplexity Deep research to try to infer what services the company might need based on it's probable current challenges.
In this case it was: "Cloud infrastructure support to manage rapid scaling, pivoting, and new product integrations as part of business model transformation[3]., Specialized consulting (e.g., industry-specific financial regulations, fintech compliance, and go-to-market strategy) to facilitate entry into new verticals and optimize operational resilience[2]."
The first site (Idealista) I tried it on flagged and blocked me / my home IP as a bot within 10 seconds.
Dead Comment
second, how does one increase the context window without requiring obscene amounts of RAM? we're really hitting the limitations of the transformer architecture's quadratic scaling...
[1] https://research.google/blog/chain-of-agents-large-language-...
Including translated index of government forms[0]
I moved to Spain a few years ago and love it, except for the paperwork. I’m slowly learning Spanish (currently 1112 day streak on Duo) but found it hard to get the right info.
So I’m using gpt4 and perplexity to do things like translate local news, government forms, and the official government update feed.