Field notes№ 01 to 03
Notes on AI and code.
RAG, agent loops and algorithms, with examples you can try as you read.
- № 01
How RAG finds the answerA model needs the right information before it can give a useful answer. Try changing what gets retrieved and see what happens. - № 02
Inside an agent loopFollow a support request through tool calls, an error and a retry. Then change the limits and see where the loop stops. - № 03
Binary search, one step at a timeStart with a guessing game, then follow the code. See why halving the search works and what happens when the data is not sorted.