GraphGeeks Podcast

Practical Trends in Graphs and AI with Paco Nathan

25 snips
Sep 24, 2025
Join Paco Nathan, an experienced AI and graph consultant, as he shares insights on the latest trends in AI and graph technology. He discusses how companies are shifting from basic code generation to real-world applications like fault detection. Paco elaborates on the rise of hybrid AI, combining neural and symbolic systems for better model efficiency. He also highlights practical workflows with tools like MLflow and DSPy, and touches on the growing interest in probabilistic graph approaches. A must-listen for anyone curious about the future of tech!
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Pragmatic LLMs Over Pure Code Generation

  • Databricks is pushing pragmatic LLM uses across the software lifecycle rather than grand code-gen promises.
  • This shift opens space for graphs to link hypotheses and evidence for tasks like fault detection and root-cause analysis.
ANECDOTE

Agent Builders Felt Like A Maker Fair

  • The AI Agent Builders World Tour felt like a high-school science fair with packed attendance and grassroots demos.
  • Paco attended with a Make Magazine editor and saw many hands-on, low-fi agent demos that impressed him.
INSIGHT

Neurosymbolic Methods Are Returning

  • Neurosymbolic and probabilistic approaches are resurging to combine symbolic reasoning with neural models for explainability and efficiency.
  • These hybrid methods can constrain LLM outputs with external symbolic systems instead of expensive internal guardrails.
Get the Snipd Podcast app to discover more snips from this episode
Get the app