Tony Xavier
Cofounder @ talentpluto, Graduate Student Researching innovations in RAG
Actions
Tony Xavier is a Graduate Student at DePaul University studying CS with a concentration in AI/ML. His previous background includes Physics and Mechanical Engineering from University of Michigan and has worked with companies such as Ford, Hyundai and PG&E.
Tony is a serial entrepreneur and has built a no-code editor, the smallest portable (foldable) electric skateboard and loves Physics.
Links
Area of Expertise
LightRAG and Voice Agent Applications
Voice interfaces are finally practical, but most RAG pipelines still depend on flat vector search. That limitation leaves callers waiting while systems fumble multi-hop context such as “HOA bylaws → city ordinance → short-term-rental clause.”
In this session, Tony Xavier will show how LightRAG, a dual-level retrieval framework that pairs dense embeddings with a Neo4j knowledge graph that will solve that gap and power real-time voice agents for property tech workflows.
You will follow three concrete stages:
Problem framing: Why single-vector RAG fails on multi-entity questions (“lease clause → city code → pet policy”).
Core concept: LightRAG’s local + global scoring—dense chunks for speed, Cypher paths for reasoning.
Minimal tool chain: STT → LightRAG (FAISS + Neo4j) → GPT-4o → TTS, all wrapped in a FastAPI service.
By the end, you will understand the LightRAG architecture, the Cypher patterns that enable relational recall, and a reference stack that will let you build your own Neo4j-backed voice agent in a weekend.
NODES 2025 Sessionize Event Upcoming
Tony Xavier
Cofounder @ talentpluto, Graduate Student Researching innovations in RAG
Links
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top