Hyperlambda
N8N recently published a number they are presumably proud of: 2.1 million lines of code produced in the last year.
That is 1,100 employees. A $5.2 billion valuation. And 2.1 million lines of code.
Last year I produced 6.2 million lines of code.
Alone.
I made a video about it, because of course I did.
The math they won't like
This is napkin arithmetic, so get your napkin out and check me.
6.2 million divided by 2.1 million is roughly three. One person out-produced their entire company by a factor of 3x. Not their backend team. Not their platform group. The company.
Now be generous to them. Assume that of those 1,100 employees, about 300 actually write code — the rest do sales, marketing, HR, and whatever else a $5.2 billion company fills its floors with. That puts the average N8N developer at about 7,000 lines for the year.
7,000 versus 6.2 million. That makes me roughly 900x as productive as their average developer.
Per person. Per year. By their own published metric.
The part that should actually worry them
Here is the detail that turns this from a flex into a problem for them: those 6.2 million lines were not even my main job.
Most of that code is training material for the Hyperlambda Generator — the custom LLM that writes Hyperlambda with a vocabulary it cannot hallucinate its way out of. The 6.2 million lines are a byproduct. A side effect of building something else.
And they came in addition to everything else that shipped in the same period: the entire dashboard rewrite, the MCP server integration, and every other improvement that landed in Magic Cloud over the last year.
Out-producing a $5.2 billion company was the warm-up.
Purely logically...
Now for the philosophical section, because the numbers invite it.
N8N is valued at $5.2 billion. By their own published metric, they produced 2.1 million lines of code last year. Simple division: roughly $2,476 of valuation per line of code.
I produced 6.2 million lines.
Purely logically, does this imply I'm worth 15.3 billion dollars?
Of course not. And that is the interesting part. Because if the conclusion is absurd, one of the inputs is absurd — and my line count is the verifiable one. It is sitting in public repositories, timestamped, one commit at a time.
So which input is the absurd one? You have two options.
Option one: lines of code mean nothing. Fine — but then why did a $5.2 billion company publish theirs as a brag?
Option two: the valuation means nothing. Also fine — but then what exactly are those 1,100 salaries resting on?
Pick whichever you like. Both answers are funny, and both answers are bad for N8N.
I'll let the VCs do their own due diligence. I just do arithmetic.
"But LOC is a terrible metric!"
Correct. It is. Lines of code is a famously stupid way to measure productivity, and I would never choose it.
I didn't choose it. They did. They published it, presumably as something to be admired. I am simply measuring myself with their yardstick.
If the yardstick embarrasses them, they picked it.
The punchline
But then again, it's not really a fair competition, innit?
You see, they're using N8N.
I'm using Magic Cloud. 😇
Come take the leverage
If one person on Magic Cloud out-ships a $5.2 billion company by their own metric, imagine what your actual team does with it. The economics are not subtle — I have measured them, repeatedly.
Magic is MIT-licensed and open source — the repository is at github.com/polterguy/magic, with documentation at docs.ainiro.io.