Memory & RAG
Threads that remember, token-aware context, hybrid search, knowledge graphs and retrieval-augmented agents.
+++ 7 recipes +++
Conversation Memoryeasy5 minYour chatbot forgets the user’s name between messages. You want every turn of a thread remembered.Token-Aware Contextmedium10 minYou assemble prompts yourself and need the most recent history that fits a token budget, with the system prompt always included.Hybrid Searchmedium15 minKeyword search misses paraphrases and vector search misses exact terms. You want both, weighted.Knowledge Graphadvanced20 minYou want facts about people, places and organizations pulled out of text and stored as a graph you can traverse.RAG Pipelinemedium10 minYou have a folder of documents and want to ask questions against it.Chunking Strategieseasy5 minRetrieval quality depends on how documents are split. You want to compare the splitters on your own text.Agent with RAGmedium10 minInstead of stuffing documents into the prompt, the agent should search the knowledge base when it needs to.