LLMs 8
- Views about AI models formed 4-6 months ago should be reconsidered.
- The biggest source of confusion about whether AI models are good or bad is that people judge them from wildly different codebases, languages, teams, and usage levels.
Thorsten Ball
Co-founder of Amp and co-host of the Raising an Agent podcast. Wrote Writing an Interpreter in Go.
Thorsten Ball did not write this page.
We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.
What Thorsten Ball says most about
software quality 8
- Anyone who believes agents no longer need Rust because they handle memory allocation well must also believe aspects of testing become useless.
- Questions whether software that works, performs well, has no bugs, is understood, and can later be modified by an agent should be called slop.
developer tools 6
- AI models are now much more than text-to-code converters.
- The terminal is dead; most developer tooling including shells, text editors and CLIs will be washed away by tokens and no longer used by humans.
coding agents 5
- An entity that can write a 1000-line program without any mistakes or compile errors may not need the same crutches human programmers do.
- Anyone who believes agents no longer need Rust because they handle memory allocation well must also believe aspects of testing become useless.
design 4
- The PM/Design/Eng triad will disappear as senseless, and Agile and SCRUM are dead.
- Much UI exists because software cannot understand user intent and serves as a human API to dumb machines, so smarter machines need far less UI.
simplicity 4
- Unnecessary code is a burden that makes a codebase harder to change
- Simple, straightforward solutions can also be elegant, though not every elegant solution is straightforward.
AI agents 3
- Future importance will lie in runtime behavior and its legibility to agents.
- Waiting for autonomous agents is not a productive strategy
AI and human skill 3
- There is no proof good code will matter in the future, since the notion is based on human ease of use and humans will not modify most code.
- The craft of writing code will disappear, while the craft of building software becomes more important than ever.
AI and jobs 3
- Some people will be priced out of producing software for the next few years if they cannot get enough tokens, putting them in a second league.
- It will take a generation for new software to replace old software, and people will still be employed writing code in 10 years.
innovation 3
- Views about AI models formed 4-6 months ago should be reconsidered.
- Tokens are the new computing paradigm on which everything will be remade as computing enters a post-binary era.
Ruby 3
- A programming language functions as a lens that shapes how a problem is perceived when solving it.
- Programming languages shape how programmers think, similar to how spoken languages shape thought.
TypeScript 3
- A programming language functions as a lens that shapes how a problem is perceived when solving it.
- Programming languages shape how programmers think, similar to how spoken languages shape thought.
writing 3
- Discomfort or effort in typing that makes someone avoid writing things down is a genuine bottleneck that should be fixed, not dismissed.
- The effort typing requires, not raw typing speed, is what determines whether it becomes a real obstacle to writing.
agency 2
- Growing conviction that people who do best with AI agents are those good at shipping things in small pieces.
- Those who love stagnation seem to have gotten enough of what they wanted, and hopefully this is an inflexion point for breaking free from its oppressive rule.
optimism 2
- This technology will create little moments of delight, whimsy, and value in everything people touch.
- Those who love stagnation seem to have gotten enough of what they wanted, and hopefully this is an inflexion point for breaking free from its oppressive rule.
remote work 2
- Face-to-face meetings are not inherently better than other forms of communication and can waste time just like anything else.
- In asynchronous, text-based communication, the loudest person in the room loses influence and it shifts to the best communicators.
software performance 2
- Using a dependency carries ongoing maintenance costs, not just upfront convenience
- Performance-critical human contributions will remain an edge case, and in 99% of software a faster algorithm or better data structure does not matter.