Skip to content

Blog

Cheap code makes engineering discipline worth more

Generating code got cheap, and anyone can do it. Engineering systems that hold up in production is the skill that gained value.

Written by

Specification and review

Published: 5 March 2026 Updated: 1 min read

The price of writing code just collapsed. The value of engineering it didn’t.

The unit economics of our industry have changed for good.

The industry splits in a K-shape

The industry is splitting into two groups:

  • Model-first companies: lean, fast, operating on razor-thin margins, shipping in weeks what used to take quarters.
  • Everyone else: still debating whether to allow Copilot.

What is already happening

  • Non-developers build production apps at Cursor meetups. The knowledge barrier is collapsing.
  • Smart engineers leave companies that ban AI. They won’t risk career irrelevance for corporate comfort.
  • Per-seat SaaS models become indefensible as teams shrink.
  • Org charts flatten. Deep management hierarchies compress into small teams backed by AI agents.

What it means to be a developer now

“What does it mean to be a software developer when everyone can develop software?”

Your language-specific, stack-specific identity is no longer a moat. When code generation is this cheap, the ability to write code is the wrong differentiator.

Generating code is easy, engineering systems is not

Here is what most analysis misses. Everyone can generate code now. Almost nobody can engineer systems.

The vibecoded SaaS breaks in production. The AI-generated authentication leaks data. The “just ship it” prototype becomes a security incident.

The real moats in this era:

  • Distribution and brand. Relationships still close deals.
  • Architectural discipline: owning whether the system works together.
  • Context engineering: structured knowledge that turns generic AI into production-grade output.
  • The engineering layer that separates prototypes from products people trust.

Cheaper development raises the value of engineering

Cheaper development makes engineering discipline more valuable, not less. When everyone can produce code, the advantage shifts to teams that produce the right code, in the right architecture, with the right quality.

The threat isn’t AI. It’s vibe coding without engineering.

Is your team built for the world of cheap code, or still pretending it’s not coming?

Further reading

  1. AI writes your tests, but not the one for the right problem A suite at 94% coverage, every test green, for a feature no user completed. The test that checks the business need depends on context only humans can supply. 2 min readShort post
  2. AI makes code easy, and only a spec keeps it simple Conversational coding piles every instruction onto the last until the code tangles. Rich Hickey's distinction between easy and simple explains why, and a research, plan, implement process is the fix. 2 min readShort post
  3. Do the sprint planning work before the planning meeting Most of what a planning meeting answers needs data, not a meeting. Agents split stories, map dependencies, order tasks, write evil user stories and check acceptance criteria before anyone sits down. 3 min readShort post

All 23 posts on Specification and review

ClientsClient work

Test the constraint in your system

The Factory Readiness Assessment connects this research to your value stream, your data, and the next decision.

Training on this topic: Agentic Engineering for Developers and QA
See the assessment

Want to check one team first? The self-check: twelve yes or no questions, three minutes, instant result, no account. Take the self-check