How to survive as a developer in the exponential age of AI
summary
Shows a concrete stack: a central prompt service that calls Gemini and other models, uses the new Responses API and experiments with MCP servers, and plugs model calls into product and invoice workflows. Teams auto‑generate content, scrape competitor prices, and built an invoice agent that multiplies throughput from ~11–12 invoices/day per person. Pitfalls include hallucinated libraries (eg “just mock window”), exposed API keys, rising costs despite current daily spend under $100, and growing technical debt despite 50–80 releases per day.
AI-written summary · about
chapters (18)
- 0:00AI hype cycle
- 3:13Software complexity and AI's role
- 7:03VIP coding pitfalls
- 11:37Software quality paradox
- 16:16AI coding pitfalls and risks
- 20:26Introduction
- 22:32Beyond typing code
- 24:16Encapsulating AI models
- 26:22Invoice recognition with AI
- 28:30Web data scraping for pricing
- 30:36New products from AI API
- 33:06AI-driven campaign creation
- 37:17AI dominance and ethics
- 41:12AI code risks
- 45:23Autonomy and team rules
- 50:20Abandoning old processes
- 53:09Small steps and automation
- 58:04Summary and AI future
related talks
26mThe Developer's Roadmap for the wave of AI Powered Applications
51mCode Smarter, Not Harder: AI-Powered Dev Hacks for All
59mAI is Eating the Developer Career! Wait, what? Maybe it's the other way around?
40mA Developer’s Adventures in AI Wonderland
58mFrom AI Sceptic to AI Sceptic Who Ships
18mDispatch from the Future: building an AI-native Company
57mOne Team’s Journey to Using AI Without the Hype
11mStop Ordering AI Takeout A Cookbook for Winning When You Build In House