My team tried it, my conclusions were:
- it took just as much prompting as a using an LLM to get simple tasks done
- took many hours to come to wrong conclusions for simple problems
- the examples in their site are definitely cherry picked
Thanks for the insights. I didn't think it would do better than other LLM but their Github issue stats were quite impressive. It's nice to see some real experience with it so we know its actual capability.