Series · Part 2 of N

Deep Due: Part 2 - Teaching the Graph to Think

X May 2026 LangGraph Python Agents LiteLLM Vector Databases AI Engineering

Series · Part 2 of N

This is a longer one than usual so grab a coffee. It's the first in a multi-part series walking through building an agentic UK corporate fraud investigator end to end. Part 1 can be found here.

All the code lives over at github.com/DaveAhearne/DeepDue if you want to follow along.

A little background

Next time...

About the author

David A'Hearne

ML Engineer and founder with 13 years building production systems, the last two focused on applied ML. I've shipped NLP scoring pipelines, RAG systems for clinical code retrieval, and LLM-integrated APIs with production MLOps tooling.

Recently completed a Cambridge Data Science Career Accelerator, and continuing towards a BSc in Mathematics. Now working as a Data Engineer in the ML team at RealityMine, Manchester.

Python ML & NLP MLOps Go Cloud Architecture

GitHub LinkedIn