What if you could have one more phone call with someone you’ve lost?
Brian Dema and a 20-person team spent six months turning 17,500 pages of his late father’s writing, client work and archives into RichieBot: a cloned voice, an Unreal Engine face and a 4-foot interactive hologram. Then a Stanford psychologist tested how close it came to the real man.
Brian Dema
Founder, Department of Growth
Founder of Department of Growth, a fractional marketing agency for fast-growing startups. Led the six-month build of RichieBot, an AI version of his late father, J. Richard Dema.
Follow Brian on LinkedIn →
Anirudh Narayan
Co-founder & Chief Growth Officer, Lyzr
Co-founder & Chief Growth Officer, Lyzr. Host of How We Agent. He built the agent that pitched Lyzr to its Series A investors.
Follow Ani on LinkedIn →Must watch if you’re building human-like AI and want to see what happens when it has to be one specific person.
Chapters
Orange dot marks a key momentFive moments worth pressing play for
Short on time? Pick up where you want.
-
Moment 01
How RichieBot was built
Six months, a 20-person team and five filing cabinets of paper:
- Scan 17,500 pages of writing, client work and archives
- Build a RAG system over a lifetime of work
- Write soul files on top of Claude
- Clone the voice from just four voicemails
- Build an abstract face in Unreal Engine
-
Moment 02
Soul files
How to write a personality for an AI without turning it into a caricature, and why “he was nice” tells a model nothing.
-
Moment 03
Testing the bot against the real man
A Stanford psychologist measured RichieBot on:
- Big Five traits
- Schwartz values
- Primal world beliefs
It came out as a “larger than life” version of his father, and Brian explains why that might be okay.
-
Moment 04
What the things we keep say about us
Why an LLM’s reading of the articles his father kept captured his personality more accurately than human memories did.
-
Moment 05
Don’t build it while you’re grieving
Brian’s advice for anyone thinking of building one for their own family, and why AI versions of people may become normal in homes and museums.
Up next
All episodes


A Breakdown of What Cisco’s AI Infra Stack Looks Like
Raj Rikhy, Senior Director of Product Management
Latest
How the US DOT used AI agents to analyze 120,000 public comments
Pavan Pidugu, Former Chief Digital and Information Officer
Want more of this?
Get notified when the next episode drops, read case studies or get AI agent use cases and templates right in your inbox every month.
No spam. Unsubscribe anytime.
