Building a RAG pipeline

Compression-based agent memory beats graph-based agent memory on the LongMemEval benchmark on the same model, and four independently-built coding agents converged on the same compression pattern.

Self-hosted graph memory is uniformly not production-ready — Graphiti has an undocumented async conflict, Mem0 OSS graph is OpenAI-only despite 47k stars, and the MCP memory server corrupts files under concurrent writes.

ChromaDB's single-node RAM-bound architecture makes it a prototyping tool with a hard scaling ceiling — LRU eviction fails, collection deletion doesn't free memory, and connection leaks cause container OOM.

Every major provider advertises mathematical or near-guaranteed schema adherence, but each implementation has undocumented complexity thresholds, ordering sensitivities, and silent failure modes that make the guarantee conditional in ways the documentation does not disclose.

Three silent failure modes in RAG pipelines — GraphRAG entity deduplication corruption, LangGraph edge routing data loss, and tool output leakage at step caps — all go undocumented by the frameworks.