Xavier Francuski
AI Researcher
Xavier Francuski contributes to the Cognee blog, with articles in deep dives, integrations, fundamentals, tutorials, cognee news. Explore their published work below.
Published topics
Deep DivesIntegrationsFundamentalsTutorialsCognee News

Published articles (39)
Agent Workflow Memory: How AI Agents Plan, Act, and Remember
Learn what agent workflow memory is, how it differs from workflow state, and how to add memory to AI agent workflows with cognee, an open-source memory tool.
Read article ↗Why AI Agents Forget and How to Fix Their Memory
Learn why AI agents lose context and how to fix agent memory with better state, retrieval, long-term storage, and cognee's memory lifecycle.
Read article ↗What Is Agentic RAG? How It Works and When to Use It
Agentic RAG puts an agent in control of retrieval — planning, choosing tools, and retrieving again when evidence is missing. Learn how it works and when it's worth it.
Read article ↗Give Claude Code Persistent Memory With cognee
Learn what Claude Code already remembers between sessions, where the gaps are, and how to set up and verify cognee's plugin for persistent, searchable memory.
Read article ↗GraphRAG vs RAG: Key Differences and How to Choose
Compare GraphRAG vs RAG across retrieval, data structure, cost, use cases, and limitations. Learn when vector RAG is enough and when graph retrieval is worth the added complexity.
Read article ↗What Is an AI Knowledge Graph? Entities, Relationships, and Use Cases Explained
Learn how AI knowledge graphs connect entities and relationships to improve search, question answering, generative AI grounding, reasoning, and agent memory.
Read article ↗cognee Joins UC Berkeley Xcelerator's 2026 Agentic AI Cohort
cognee has been selected for Berkeley RDI's Xcelerator 2026 Spring Cohort, a non-dilutive program for agentic AI startups, alongside Narada AI, RELAI, and Headroom.
Read article ↗What Is GraphRAG? Retrieval-Augmented Generation with Knowledge Graphs Explained
GraphRAG adds a knowledge graph to the RAG pipeline so retrieval can follow relationships instead of returning isolated chunks. Learn how the pipeline works, when to use local vs global search, and where GraphRAG earns its complexity over standard RAG.
Read article ↗What Is RAG? Retrieval-Augmented Generation Explained
RAG pairs retrieval with generation so an LLM can answer from external knowledge instead of just its training data. Learn how RAG works, what it solves, and where chunk-based retrieval starts to hit its limits.
Read article ↗LLM Hallucination Solution: How to Reduce Wrong AI Answers
Grounding is the most effective LLM hallucination solution: make the model answer from retrieved, verifiable facts instead of training memory. Retrieval quality, structured knowledge, verification layers, and feedback loops all determine how many wrong answers survive to production.
Read article ↗LLM Hallucinations: What They Are & How to Detect Them
LLM hallucinations are fluent, confident-sounding answers that are false or unsupported by any source. Learn what causes them and the detection methods — groundedness checks, self-consistency, LLM judges — that catch them before users act on them.
Read article ↗AI Knowledge Base: Building a Retrieval-Ready Knowledge Layer
An AI knowledge base needs more than stored documents. Source context, entity relationships, and plural retrieval strategies are what turn stored data into reliable, agent-ready knowledge.
Read article ↗