Beyond Coding

What Coding Agents Need From Engineers to Be Trusted

14 snips
Nov 11, 2026
Fausto Albers, an AI consultant at WonderWhy, builds production software with coding agents. He explores why models are absorbing their scaffolding, how engineers can design reliable agent loops, and what agents need before working autonomously. They also discuss observable quality gates, layered memory systems, disposable versus durable software, domain expertise, and AI’s unsettling impact on cybersecurity.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Intent-Built Software Could Disrupt SaaS

  • If software can eventually be generated directly from intent, customers may build what they need instead of buying centralized SaaS.
  • Fausto Albers believes this could fundamentally weaken the SaaS model within the next decade.
ANECDOTE

The Restaurant Recommendation Startup That Challenged Preference Data

  • Fausto Albers built a restaurant recommendation startup to reduce the time customers spent ordering through phones during COVID-era restrictions.
  • He initially assumed preferences could be perfectly matched through quantified flavor vectors and recommendation data.
ANECDOTE

Why Fausto's Restaurant Chatbot Failed

  • Fausto Albers launched an early generative-AI restaurant chatbot before function calling existed, but the project failed despite strong recommendations.
  • He concluded that hospitality needs human interaction, while AI can augment the bartender's knowledge and personalization.
Get the Snipd Podcast app to discover more snips from this episode
Get the app