Markus Harrer
Getting to Know Your Legacy (System) with AI-Driven Software Archeology
#1about 5 minutes
Applying archaeological techniques to legacy software systems
Legacy systems present challenges like poor documentation and missing context, which can be addressed by applying archaeological methods to understand their history and structure.
#2about 4 minutes
Using excavation to map your legacy codebase
The Wheeler-Kenyon method can be adapted to software by creating a grid-like treemap of a codebase to visualize file age and development hotspots.
#3about 10 minutes
Identifying code patterns with AI-driven typology
Typology involves classifying scattered source code files into technical and business concepts, a repetitive task that large language models can automate.
#4about 5 minutes
Scoring the conceptual integrity of software components
An LLM can score how well a piece of code implements its intended concept, helping to identify trustworthy and mixed-up parts of the system.
#5about 4 minutes
Reconstructing component history with chaîne opératoire
The chaîne opératoire technique uses an LLM to analyze commit history and generate a timeline of a component's evolution, revealing its purpose and key contributors.
#6about 1 minute
How to effectively leverage AI for legacy code
Successfully using AI on legacy systems requires breaking down the problem and providing specific context rather than feeding the entire codebase to a model.
Related jobs
Jobs that call for the skills explored in this talk.
Featured Partners
Related Videos
New AI-Centric SDLC: Rethinking Software Development with Knowledge Graphs
Gregor Schumacher, Sujay Joshy, Marcel Gocke
Data Science on Software Data
Markus Harrer
Grappling With Clunky Old Software? Start by Understanding What’s Inside!
Luc Perard
Leapter: The Reinvention of Software Development? A Future Built On AI Generated Code.
Robert Werner
AI-Powered Code Documentation: Simplify the Complex
Patrick Schnell
Livecoding with AI
Rainer Stropek
Migrating from COBOL with AI: A Moonshot Demo
Julia Kordick
Leveraging Large Language Models for Legacy Code Translation: Challenges and Solutions
Michael Niebisch
From learning to earning
Jobs that call for the skills explored in this talk.


Senior Backend Engineer – AI Integration (m/w/x)
chatlyn GmbH
Vienna, Austria
Senior
JavaScript
AI-assisted coding tools




Part Time Junior Python Backend / GenAI Support Intern
Eltemate
Amsterdam, Netherlands
Remote
Junior
Python
GraphQL
Agentic AI Architect - Python, LLMs & NLP
FRG Technology Consulting
Intermediate
Azure
Python
Machine Learning
Security-by-Design for Trustworthy Machine Learning Pipelines
Association Bernard Gregory
Machine Learning
Continuous Delivery
Product Leader in AI-Powered Software Modernization
CAST
Canton de Meudon, France
Senior
Agile Methodologies




