Core Memory

The Aussie Man Who Used AI To Create A Cancer Cure For His Dog

49 snips
Mar 18, 2026
Paul Conyngham, an AI and machine learning practitioner who used computational tools to design a personalized mRNA cancer therapy for his dog Rosie. He explains sequencing Rosie’s tumor, using AI to identify targets and design an mRNA vaccine, coordinating lab work and approvals, and the mixed public reaction and regulatory questions this breakthrough stirred up.
Ask episode
AI Snips
Chapters
Books
Transcript
Episode notes
ANECDOTE

Self-Taught Pipeline From Misdiagnosis To Sequencing

  • Paul Conyngham taught himself cancer basics and used LLMs to map treatments after his dog Rosie was misdiagnosed for 11 months.
  • He sequenced Rosie's tumor and blood, compared genomes, and learned immunotherapy mechanics via ChatGPT and notes in Obsidian.
INSIGHT

Multimodal Immunotherapy Was Key

  • Conyngham used a multimodal therapy strategy: mRNA vaccine as the targeted 'sniper' plus PD-1 inhibitors and tyrosine kinase inhibitors.
  • The vaccine trained T cells while PD-1 blockade prevented tumor-induced T-cell shutdown and TKIs starved tumor blood supply.
ANECDOTE

From Raw Reads To Protein Models And Drug Designs

  • Paul assembled Rosie's raw sequencing data (300 GB per sample), modeled C-Kit in AlphaFold2, then ran docking and genetic algorithms to find strong binders.
  • He abandoned novel compound synthesis because it would require full drug development and regulatory trials.
Get the Snipd Podcast app to discover more snips from this episode
Get the app