Does AI widen the productivity gap between top engineers and everyone else?

Edwin Chen · Jul 21, 2025 · hedged

AI coding tools disproportionately amplify already-strong engineers rather than lifting average ones, and this will be even more true in future

Strong engineers have far more ideas than they have time to implement; AI removes the drudgery of day-to-day coding so those ideas can be put to paper

Scope: may be true of both groups today; more pronounced in the future; assumes AI is better at removing drudgery than generating the best ideas

12:26 20VC: Scaling to $1BN+ in Revenue with No Funding: Surge AI | The Most Insane Scaling Story in Tech |

Zach Lloyd · Oct 17, 2025

AI coding tools counterintuitively favor higher-quality, more senior engineers and actively cause problems for junior or unsophisticated ones

Less sophisticated engineers get code they don't understand, that can't ship, won't pass code review, or contains security bugs; using the tools successfully requires being more sophisticated than the agent

Scope: professional environments; paradox: senior developers often don't want to use the tools because they feel they can do the work themselves

18:39 20VC: The Startup Adding $1M ARR Every Week | Competing Against OpenAI's Codex and Claude Code: Who Wins | Why Gemini is Failing and GPT-5 Is Winning | Do Margins Matter in a World of AI | The Ugly Truth About AI Coding with Zach Lloyd, Warp

Mike Cannon-Brookes · Oct 13, 2025 · hedged

AI tools shift the whole distribution rather than compressing it: yesterday's 10x engineer becomes a 100x engineer and the average engineer becomes a 10x engineer

Performance comes down to the person's learning style, ability to ingest information and how they solve problems, which new tools amplify rather than equalise

31:37 Tools shift the whole distribution up preserving relative gaps rather than widening them

Anton Osika · Aug 18, 2025

AI's multiplier depends on the work: it can take a junior engineer from zero to one on things they couldn't do, but on complex systems requiring years of context it does nothing for the junior while taking a 10x engineer to 100x

AI can bridge a specific skill gap, but it cannot supply the system understanding a junior lacks

35:19 Task complexity determines whether ai bridges junior gaps or multiplies senior output

Amjad Masad · Mar 10, 2023 · hedged

Average developer salaries may fall while the most creative and hardworking developers earn far more

AI gives the best engineers enormous leverage — Carmack is currently limited by his typing speed, but directing thousands of AIs at a higher level of abstraction removes that constraint and automates the menial work

29:09 Average developer pay falls while top developers earn dramatically more due to ai leverage

Paul Erlanger · Jun 27, 2026

AI makes the very best people more valuable, not less, because it absorbs the lower-level work while taste and direction remain human

You can use ChatGPT to make art but you still need creative direction; engineers become architects while AI does what a B-tier engineer would have done

24:05 Ai absorbs b tier work so taste and direction gain value

Matan Grinberg · Jun 13, 2026

The bifurcation of engineering talent is directionally right, but the right frame is 'load-bearing individuals' rather than 10x/100x output multipliers

10x implies pure output like lines of code, which is meaningless now that anyone can generate a billion shit lines; what matters is whether removing a person makes things fall apart

7:04 Bifurcation is real but measured by load bearingness not output multiples