Problems.
Write the piece, drive the whole agent loop, or repair a repo by prompting a coding agent. Most are graded against a real model, because that is what you will ship against.
0 of 85 solved
Show me at , . 85 matches
- 61Carry the Conversation YourselfdebugagentL01easy
- 62Ask Again When the JSON Is WrongimplementagentL01medium
- 63Handle the Reply That Got Cut OffimplementagentL01medium
- 64Price a CallimplementstdlibL01easy
- 65Assemble a Reply From a StreamimplementstdlibL01medium
- 66Sample the Next TokenimplementstdlibL01medium
- 67Fix the System Prompt That Leaksdebug · liveagent · liveL01medium
- 68Pick the Examples That Fit the Questionimplement · liveagent · liveL01medium
- 69Compact a Long Conversation Without Losing the Decisionimplement · liveagent · liveL01medium
- 70Put the Answer Where the Model Will Read Itimplement · liveagent · liveL01medium
- 71Lay Out the Prompt So the Cache HitsoptimizestdlibL01medium
- 72Search by Meaning, Not by Wordsimplement · liveagent · liveL02medium
- 73Rerank With a Model That Understands the Questionimplement · liveagent · liveL02medium
- 74Check Every Sentence Has a Sourceimplement · liveagent · liveL03medium
- 75Rewrite the Query Before You Searchimplement · liveagent · liveL02medium
85 problems, showing 61 to 75