Ask HN: What will humans do when AI writes and reviews code?

2•pomarie•1d ago

I’m noticing more and more teams are using cubic (an code reviewer – I’m the founder) to review PRs written by AIs (eg. Devin, Codex…).

Some teams already run an entirely AI-driven pipeline where AI writes the code, reviews it, and humans just click Merge at the end.

Interestingly, the AI reviewer often finds errors in AI-generated code. For now, AIs can review AIs code with OK results – probably better than a junior-mid level dev.

It got me thinking of the future…

1. Junior developer growth*:* Traditionally, code reviews are how junior developers learn. Have you noticed changes in how they learn since codegen has become a thinh? 2. Society expects self-driving cars to be dramatically safer (e.g. 10x better) before accepting them. Do you expect similar standards from AI code reviewers, or is slightly better than human good enough? 3. As AI surpass humans at writing and reviewing code (at least for technical correctness), how do you see the role of code reviews changing? 4. Philosophically, do you think code generation AIs will ever become so effective that specialized AI code reviewers won’t find any issues?

Comments

TheMongoose•1d ago

Charge a lot more to come in and re-write the spaghetti mess that the vibe coder left the business with into a product that actually works.

pomarie•1d ago

Circle of life!

coffeecoders•1d ago

Just like IDEs with autocomplet, auto generators (getter/setter etc.), CI/CD pipelines, and Docker changed how we build software, LLMs code generators and reviewers will be another tool to transform our role and boost productivity. It will handle the repetitive parts so we can focus on higher-level thinking.

sebnado•1d ago

We’ve been running an AI-first dev loop in production for ~2 years (disclaimer: I help build Ze1 and Sandscape, they are both Ai driven products). A few things we’ve learned:

Instead of cranking boilerplate, they spend Day 1 reviewing AI diffs. We pair them with a senior for a “why did the agent choose this?” teardown.

What matters is mean-time-to-rollback. If the agent + test harnesses catch breakage faster than a human pair can, 2 × better is already good economics. Reliability engineering beats perfection thresholds.

Syntax, style and most unit-level bugs are now linted or auto-fixed. Humans zoom out to architecture, data contracts, threat models, perf budgets, and “does this change make sense for the product?”. So even juniors are now a lot more involved on the sujective elements of development

So I think that as the easy bugs vanish, new failure modes show up: latency cliffs, subtle privacy leaks, energy use, fairness. The goal posts moves

Making Satellite Calls with a 1973 Telephone (and Inmarsat Mobile Terminal) [video]

Biotech founders and drug hunters who were unlikely to succeed, and yet they did

Just how bad are we at treating age-related diseases?

I Went Inside the Factory for the Insane Printer Everyone's Talking About [video]

Hyperstition

Uberization of Knowledge Work

Why I Wrote the Beam Book

Show HN: RSS Feed Validator and Analyzer

Schleswig-Holstein's Bold Open Source Leap: A Model for Digital Sovereignty?

Where are all the data centres and why should you care?

DeepSeek may have used Google's Gemini to train its latest model

The Goodness Advantage that we founders have

How Computational Imaging Is Revolutionizing Astrophotography

A Fatal Tesla Crash Shows the Limits of Full Self-Driving

Secure Minions: private collaboration between Ollama and frontier models

Atomic Secrets [video]

Trump memecoin wallet in 'absolute chaos' as family org unaware of launch

Hypervisors for Memory Introspection and Reverse Engineering

Trump-linked asset manager files Truth Social Bitcoin ETF with SEC

Creating cross-compiled Docker images from Go binaries

Search Params Are State

I Accidentally Became Responsible for Shopify's Internal AI Chat

Show HN: Anything API – JSON response for any request

Nvidia 2026 Q1 Financial Results

Show HN: Generate SVG Using AI

Deleting X: Why Sigdoc Left the Platform – Communications of the ACM

Show HN: Chonkify – updated 1kB chunker, now grapheme-aware

Ask HN: Is learnig just pattern matching and project known patterns to find new

Show HN: I built an AI girlfriend – no back end required

The (paid) macOS music player that Apple Music should have been