Ladybird consumes next to no resources beyond donated programmer time. It might take 1 year to complete, or 50. It might pause for periods of time if there is no programming happening.
A search engine is fundamentally simple software. But the costs to make it work are enormous. Scrape the whole web, continuously. Build servers big enough to cope with request traffic, store the index, etc. You're talking huge amounts of money in hardware, massive networking bills, an army of devops to keep it running.
Then you'll need engineers to continually fight SEO, deal with spam overload and so on.
The fact that the source code is open source or not is irrelevant. It's the least important and least expensive part of the puzzle.
But to be a bit ontopic also: I want a search engine to be unbiased, without any opinion.
Currently there are no such engines: Google and Bing are too political, and they favor very heavily a small set of results. Other search providers are usually too niche, either intentionally (e.g. IIRC Marginalia is intentionally indexing only small sites, and they are filtered too), or not intentionally (e.g. small team with not enough hw). These engines might give less opinionated (or give a different opinion) results for a small subset of queries, but they are rarely usable as general engines.