Their marketing language is misleading. They must still use some transformer language model backbone to encode the text input (BERT or decoder-only LLM). The biggest difference is the output, instead of auto-regressively generating tokens, they produce probabilities over a bounded set of decisions (more flexible classification).
Human reviews are going away for most PRs these days. If the LLM decides the risk level of the change is low, the PR can be merged without another human in the loop. That's at least the direction many companies are taking. Only require a human judgement when necessary.
A coding agent driven by a large LLM can delegate smaller tasks to a faster model. For example searching through the codebase for references, examples, or established patterns. They are treated as tools and don't pollute the main agent's context.
Why wouldn't getting more customers the plan? Anthropic doesn't acquire companies to have a lower market share. There is clearly a consolidation and a rush to get as much of the developer market as possible.
It's called problem decomposition and agentic coding systems do some of this by themselves now: generate a plan, break the tasks into subgoals, implement first subgoal, test if it works, continue.
How do you join two datasets using r-trees? In a business setting, having a static and constant projection is critical. As long as you agree on zoom level, joining two datasets with S2 and H3 is really easy.
I wouldn't say R-trees solve the problem better. Joining multiple spatial dataset indexed with r-trees is more complex as the nodes are dynamic and data dependent. Neighborhood search is also more complicated because parent nodes overlap.
That's not true when tiling the Earth though. You need 12 pentagons to close the shape on every zoom level, you can't tile the Earth with just hexagons. That's also why footballs stitch together pentagons and hexagons.