What happens after coding is solved? | Fiona Fung (Claude Code & Cowork)

21 insights saved from this video by @technology

@technology
Make new mistakes: learning from mistakes is acceptable, while zero-mistake thinking can signal excessive caution and insufficient speed.
@technology
Anthropic engineers now ship roughly eight times as much code per quarter as they did between 2021 and 2025.
@technology
AI is loosening engineering’s grip on the roadmap: product managers can sometimes implement features themselves.
@technology
AI-generated tests remove the upfront tax that made test-driven development impractical for many teams.
@technology
Build at the edge of what models almost do. When they catch up, you may lead those who waited.
@technology
AI productivity is moving past token maxing toward a harder question: what value did we create, and what did it cost?
@technology
AI can search an unstructured folder for a specific document and perform informal competitive analysis for a small business.
@technology
JIT planning keeps roughly one month of priorities visible, then tests them against reality every week.
@technology
A painful, bleeding rash on the nose returned after hotel shampoo and improved after switching to organic shampoo.
@technology
Software delivery escaped rigid CD-manufacturing deadlines when teams gained the ability to ship software online.
@technology
CD manufacturing made engineering time scarce and planning essential. Online delivery and AI-assisted coding moved the bottleneck to verification.
@technology
Knitting uses two stitches, while programming uses binary zeroes and ones.
@technology
Planning can create alignment today and become obsolete tomorrow when the surrounding environment changes.
@technology
The move from Vim and terminal debugging to IDEs brought integrated debugging, breakpoints, and multithreaded debugging.
@technology
AI-era teams need creative builders with product sense and deep experts for technically difficult areas.
@technology
AI shifts engineering ambition from unblocking small features to asking how ambitious the product can become.
@technology
Leaders using the product tell teams their work matters and leadership remains engaged, not detached.
@technology
Starting as an individual contributor gives new hires valuable time inside a company’s codebases, tools, languages, and shipping practices.
@technology
Small-business owners work brutally hard on narrow margins while spending hours on invoicing, expenses, and other disliked administration.
@technology
Engineers need product sense alongside coding ability to judge whether their work actually improves the product.
@technology
AI coding has removed the bottleneck and raised the ceiling on what one person can accomplish.

More from Lenny's Podcast

See everything from Lenny's Podcast

Post on X

Keep what you found in What happens after coding is solved? | Fiona Fung (Claude Co…

A free account saves any of these cards to your Collection, and Korva resurfaces them so you actually remember.