founder, app architect, AI engineer
🇨🇭
Joined November 2018
- Tweets2.6K
- Following59
- Followers193
- Likes981
Sebastian Fichtner retweeted
The smaller the autonomously governed jurisdiction, the more it's governed by common sense.
Sebastian Fichtner retweeted
The value of unit test seems to be very dependent on the type of functionality covered.
I totally get your argument and can see it apply in many cases.
At the same time, unit tests have recently been a total unlock for adding and measuring language support in codeface.io.
I think some factors that made unit tests valuable here are:
- core functionality easily testable
- huge matrix of repetitive functionality
- tests are a good interface for humans to verify that the intended features work
- consistency across the feature matrix is critical
- documenting what works already, what doesn't and what can't work is critical
- importantly: it's often trivial to make one cell in the huge matrix work while breaking another.
The agent broke things, even rewrote tests in an unintended way until i had the test matrix and how to work with it documented clearly.
Sebastian Fichtner retweeted
In the scenario you want to appreciate the beauty of life and the ingenuity and creative expression of humanity.
Sebastian Fichtner retweeted
Ich frage mich wie oft sie das hören musste bis sie es nachplappern konnte.
Sebastian Fichtner retweeted
Following their neurotic world view to its logical end, the mainstream has completely contorted itself into demons.
Sebastian Fichtner retweeted
Some people need to believe in the non-material human essence to feel special. It's a desperate clinging to the ancient notion of a soul.
Sebastian Fichtner retweeted
What he says,
what he intends,
what he really desires,
what he really needs,
or what he really deserves?
Pick your poison 😈
Sebastian Fichtner retweeted
🇩🇪 A STATISTICAL CURIOSITY FROM THE SAXONY-ANHALT ELECTION
AfD share of in-person votes: 51.20%
AfD share of postal votes: 24.27%
That’s a difference of 26.93 percentage points.
Using a two-proportion z-test, and assuming postal and in-person voters had the same underlying probability of voting AfD, this difference is about 278 standard deviations apart.
The probability of observing a difference at least this extreme purely by random sampling is approximately 10⁻¹⁶⁷⁸¹, or roughly 0.000…% (16,778 ZEROS before the first non-zero digit).
Also noteworthy: the share of postal voters actually fell, from 29.1% in 2021 to 27.4% in 2026, which means that the parties to the left of the AfD failed to mobilize voters to participate in the election through postal voting.
Important: This does not prove fraud. But with the postal-vote share actually declining, a divergence this extreme is statistically virtually impossible.
Sebastian Fichtner retweeted
There is one way to be puzzled which is to propose you know better, and there is another way which is to ask interesting questions.
Sebastian Fichtner retweeted
Hey @spacexai your marketing implies business SuperGrok includes the same quota as consumer SuperGrok. I feel mislead. Please sort this out and respond to emails/refund requests.
Grok Build even points me to a subscription I already have:
Sebastian Fichtner retweeted
Funny though how we, the $TSLA community, always manage to turn cautious speculation into "promise."
Elon carefully words that he THINKS SpaceX MIGHT be ABLE to reach APPROXIMATELY $1T revenue in 2030.
Then the first influencer reports it as a prediction, the second then reports it as a promise. Every single time.
Sebastian Fichtner retweeted
Endlich konnte ich FSD in Deutschland testen. Wir sind fast eine Stunde durch Dörfer, über Landstraßen und Autobahnen gefahren. Alles lief perfekt. @PSchnieder, wann retten wir Leben in Deutschland?🇩🇪
Sebastian Fichtner retweeted
Grok Build’s smartest move: It heavily uses Python.
This lifts reasoning to a higher level and provokes fewer mistakes – perfectly mixing generative creativity with deterministic reliability.
Other agents still haven’t figured this out. They run less code and write it in Bash.
Sebastian Fichtner retweeted
Success with #AI requires at least 2 of 3 factors:
1️⃣ Specialized Domain
→ and UX tailored to one niche problem
2️⃣ Unique Data
→ ideally enough to train your own models
3️⃣ Embodiment
→ ties into the real physical world
Glad to see my thesis reflected in current discourse:
My AI investment thesis is that every AI application startup is likely to be crushed by rapid expansion of the foundational model providers.
App functionality will be added to the foundational models' offerings, because the big players aren't slow incumbents (it is wrong to apply the analogy of "fast startup, slow incumbent" here), they are just big. Far more so than with any other prior new technology, there is a massive and fast-moving wave that obsoletes every new app almost as fast as it can be invented. There is almost no time to build a company and scale it.
There are two ways AI application startup founders can make money:
- Make a flash-in-the-pan app that generates a ton of cash and bank the cash (my estimate is that you have about 12-18 months cashflow generation)
- Make a good enough app that you get acquired by one of the big players for sufficient equity
The situation is highly unstable - we don't know if it's going to crash or go to the moon but both scenarios make it very unlikely that any AI application startup will independently become a generational supercompany (baseline odds are low to begin with).
The best odds are finding an application niche in a highly specialized field with extremely unique and specific data barriers, ideally ones relating to real atoms (hardware or world-related) data and not software/finance.
Sebastian Fichtner retweeted
My coworker got promoted over me.
He was worse at coding. Better at politics.
I wrote better code. Fixed more bugs. Shipped faster.
He talked in meetings. Took credit. Played the game.
He got the promotion. I got "keep up the good work".
That's when I realized: corporate rewards politics, not performance.
Six months later I quit. Started freelancing.
Now I make 3x his salary. No politics. No credit-stealing. Just solving problems and getting paid directly.
The best developers rarely get promoted. They get used.
Companies optimize for compliance and communication. Not competence.
If you're technically great but politically terrible, you'll never win at corporations.
Leave. Build your own thing. Get paid for your actual value.
Sebastian Fichtner retweeted
Das Kammergericht Berlin hat als höchste zivilrechtliche Instanz entschieden, dass die Online-Plattform Linkedin alle Beiträge löschen darf, wenn sie der Weltgesundheitsorganisation (WHO) widersprechen.
Das Gericht sagt: Es kommt nicht darauf an, ob die Beiträge falsch oder richtig seien, sondern nur darauf, dass sie im Widerspruch zur Auffassung der WHO stünden.
Das kommt der WHO sehr entgegen, denn die will ja, dass Mitgliedsstaaten "Kernkapazitäten" im Bereich Risikokommunikation vorhalten, um entsprechend mit Fehl- und Desinformation umzugehen.
Hat nur relativ wenig mit Demokratie und Meinungsfreiheit zu tun.
Irre, einfach irre.
Sebastian Fichtner retweeted
Das ist die entscheidende und eigentlich tiefe Frage.
Ich glaube die Antwort ist mehr auf der psychologischen und kulturhistorischen Ebene zu finden und weniger auf der ökonomischen oder selbst politischen ...
Es geht hier scheinbar um die Neurosen im kollektiven Unbewussten einer ganzen kulturellen Generation. Es ist so rätselhaft wie schauerlich.