Home » Burn The Map: Digital Twins, Human Judgment, and AI That Actually Works w/ Stephanie Sylvestre

2026

Burn The Map: Digital Twins, Human Judgment, and AI That Actually Works w/ Stephanie Sylvestre

In This Episode:

We talk to Stephanie Sylvestre about what it means to build AI that actually helps people instead of just adding more noise, risk, and technical debt to the pile. She’s been working in this space long before the current AI hype cycle, and her perspective is refreshingly grounded: useful AI should remove busywork, expand human capacity, and still keep a human in the driver’s seat.

Stephanie walks through the accidental origin story behind AvatarBuddy, her belief that mentorship is really a supply chain problem, and why digital twins could make guidance, support, and opportunity available at a scale the real world has never been able to offer. We also get into the less sexy—but way more important—side of AI: responsible implementation, secure systems, bloated code, fake experts, and why “faster” is not the same thing as “better.”

What We Cover:

  • How a misheard conversation led Stephanie to the idea of digital mentors
  • Why human-centered AI beats AI-for-AI’s-sake every time
  • The real reason vibe coding won’t replace actual developers
  • What most people get wrong about autonomous agents and AI safety
  • Why expertise, judgment, and ethical responsibility still matter in an AI-first world

Guest Bio:

Stephanie Sylvestre is an AI entrepreneur, operator, and co-founder of AvatarBuddy, where she’s building practical, human-centered AI agents designed to remove busywork and expand access to mentorship and support. Before that, she led technology and operations in high-stakes environments, including public-sector leadership in Miami, where her long-range cloud and automation strategy helped organizations stay functional through COVID disruption. Her work sits at the intersection of systems thinking, responsible AI, and real-world human impact.

Enjoy the episode. This show is brought to you by Wrench.ai.

Follow Dan:
LinkedIn: https://www.linkedin.com/in/danbaird/
X: https://x.com/mrdanbaird

Follow Mike:

LinkedIn: https://www.linkedin.com/in/stephanie-a-sylvestre/

Follow the Pod:
YouTube: https://www.youtube.com/@burnthemappodcast
Twitter/X: https://x.com/BurnTheMapPod
Instagram: https://www.instagram.com/burnthemappodcast/
TikTok: https://www.tiktok.com/@burnthemappodcast
BlueSky: https://bsky.app/profile/burnthemappodcast.bsky.social

Selected Links From This Episode:

People and Organizations Mentioned:

  • Stephanie Sylvestre
  • Dan Baird
  • AvatarBuddy
  • Wrench.ai
  • Hewlett Packard
  • Belize
  • Miami
  • Meta
  • Google

Show Notes & Timestamps:

  • 01:14 — The accidental origin story: mishearing “digital mentors”
  • 02:38 — Why mentorship is a supply chain problem AI can help solve
  • 03:18 — Belize, poverty, and the systems view that shaped Stephanie’s worldview
  • 04:42 — Learning computer science from an original Unix coder
  • 05:23 — From consulting to CIO, and the backstory behind “CIO of the Year”
  • 06:18 — Building a fully cloud-based operation before COVID hit
  • 08:06 — Funding agencies through the pandemic and keeping Miami from going sideways
  • 10:41 — The early days of AvatarBuddy and building with teens in low-resource communities
  • 12:33 — First prototype, intern talent, and mentoring young operators without burning them out
  • 16:29 — Auto-backup, safe failure, and how to help people push the envelope
  • 17:22 — Mining the internet for trends and developing real technical talent early
  • 19:07 — Why coding is only a fraction of what great developers actually do
  • 20:35 — AI-generated code, technical debt, and the problem with bloated repos
  • 22:34 — The right and wrong ways to use vibe coding
  • 23:16 — Security risks, root access, and why sloppy AI implementation gets dangerous fast
  • 32:57 — Open tools, security updates, and not surrendering the driver’s seat to AI
  • 35:47 — Why unsupervised agentic AI is like hiring a teenager to handle your top customer
  • 37:56 — Responsible AI, human review, and the real copy-paste-edit workflow
  • 39:53 — Why content knowledge still matters in an AI world
  • 41:13 — The future of AI: smaller models, data vaults, and specialized agents
  • 44:52 — Where humans still create the real value
  • 47:41 — How to spot true experts versus charlatans
  • 50:05 — When not to use AI, and why simple beats impressive
  • 52:51 — Stephanie’s filter: too complex, too costly, or ethically off? Hard no.
  • 53:26 — The future of AvatarBuddy: digital twins and personality chips
  • 55:37 — Trauma-informed AI, biometrics, and staying in equilibrium
  • 56:23 — Where to find Stephanie and a Burn The Map listener bonus
  • 56:45 — Who Stephanie follows to stay sharp on AI

New Episodes

Subscribe to our Newsletter!