Watch more of .local London 2026 → https://www.youtube.com/playlist?list=PL4RCxklHWZ9tH01MTlChYwUqN8Cm2tl2r Speakers: James Melvin, Principal Innovation Software Engineer at LexisNexis Risk Solutions This session explores how LexisNexis Risk is moving beyond basic AI search by building a system that actually understands how data connects over time. In this session, we’ll explore how we use MongoDB and a high-performance Rust engine to power an advanced GraphRAG-based Temporal Knowledge Graph Reasoning (TKGR) framework. We rely on MongoDB as our central hub to manage documents, map complex relationships, and verify timelines. This ensures our AI models get highly accurate, grounded context rather than just semantically similar summarized text. Join us to learn how we turned MongoDB into a high-speed, time-aware reasoning engine that drives better AI decision-making. This gives our Large Language Models exactly the grounded, reliable context they need by transforming a standard, static Knowledge Graph (KG) into a Temporal Knowledge Graph (TKG). Join us to learn how temporal graphs on MongoDB can transform RAG workflows into a more robust decision-making engine. 00:00:00 - Welcome and Session Introduction 00:01:01 - Project Origins: The Three Pillars of Advanced Knowledge Graphs 00:02:19 - Ontologies & Semantic Governance: Normalizing User Inputs 00:03:56 - The Limitations of Standalone Vectors & The Synchronization Tax 00:05:54 - Building a Hybrid Database internally on MongoDB 00:07:04 - Overcoming Document Corpus Noise and AI Hallucinations 00:08:40 - Solving Size Limits: Nodes & Relationships as Collections 00:09:10 - Grounded, Time-Aware Reasoning (The Temporal Aspect) 00:11:56 - AI Readiness, Data Normalization, and Custom ETL 00:14:40 - Searching Communities: Local Search vs. Global Search (The Leiden Algorithm) 00:17:08 - The Temporal Tuple Architecture: Subject-Predicate-Object-Time 00:19:03 - Why Rust & MongoDB Beat Python for Enterprise Compute Performance 00:22:45 - High-Level Architecture Review & The Advanced RAG Solution 00:24:22 - Tabular Facts: Managing Enterprise Data Over Time 00:26:34 - Why Text-to-SQL Falls Short for Complex Queries 00:29:10 - Bitemporal Querying: Business Time vs. Service Time 00:31:36 - Using Graph RAG Context to Drive Accurate MQL Generation 00:32:45 - Key Lessons Learned & Unified Storage Infrastructure 00:34:22 - Audience Q&A: Evaluation Process and Choosing MongoDB over Postgres 00:36:06 - Audience Q&A: The Transition to a Schemaless Solution Subscribe to the MongoDB for Developers YouTube Channel: https://www.youtube.com/@MongoDBDevelopers?sub_confirmation=1 Sign-up for a free cluster → https://www.mongodb.com/cloud/atlas/register Subscribe to MongoDB YouTube→ https://mdb.link/subscribe Visit Mongodb.com → https://mdb.link/MongoDB Read the MongoDB Blog → https://mdb.link/Blog Read the Developer Blog → https://mdb.link/developerblog MongoDB for Developers YouTube Channel → https://www.youtube.com/@MongoDBDevelopers

 400      9