Co-founder & CTO @getmimo
Vienna, Austria
Joined August 2011
- Tweets1.2K
- Following306
- Followers451
- Likes1.1K
Dennis Daume retweeted
GPT + Opus (via Copilot CLI) helped me uncover 180 JIT bugs, each with a minimal C# repro using only safe code 😶 - most had been hiding from us for many releases github.com/dotnet/runtime/is…
Dennis Daume retweeted
A good general rule for optimizing your coding loop right now is if something *feels* tedious and annoying, you should take a step back and ask yourself how an agent or simple tool an agent can build can make it less tedious.
For example, if line-by-line code review is tedious, what artifacts or proofs could you generate that would give you both understanding and confidence in a given change so that reviewing it isn't a slog?
Dennis Daume retweeted
Going to be honest...
I have almost entirely stopped using GPT since Opus 5.5. Partially because the performance is of Anthropic's new model has been so great, and partially because I never had any quota left on my Codex sub after a couple of days of usage.
Dennis Daume retweeted
I took on some tech debt yesterday for the sake of launching a customer feature asap, and then rearchitected it away 15 mins later.
No more tech debt accruing interest - the only debt is comprehension debt.
Dennis Daume retweeted
Österreich führt mit Oktober eine Paketsteuer von € 2,00 pro Bestellung ein. Steuerschuldner ist der Versandhändler.
Nun schlagen die Versandhändler die Paketsteuer einfach auf den Preis auf, aber das beste daran, nein nicht € 2,00 sondern € 2,40 weil Ust. 😂😂 Eine Steuer auf die Steuer, abgewälzt auf die Konsumenten. 😂😂 Sorry, aber das ist echt witzig 😂😂🥲
Dennis Daume retweeted
Mega 💯
Somit wird die nächste Urlaubsfahrt nach Kroatien durch Slowenien schon zu gut 70% autonom werden 😎
Im Entwicklungsland Österreich 🇦🇹 🍌 fahr ich halt noch selber …
Dennis Daume retweeted
Oh my god, Firebase’s SDK started to crash ALL iOS apps that were using it, for ALL sessions, just like that, no way for any of these apps to do anything…
How amateur is all of this from any SDK, but especially one from a company with as high of an engineering bar like Google…
github.com/firebase/firebase…
incase anyone has been wondering why their iOS apps are intermittently crashing since today morning 🙃
reminds me of the Facebook swizzling crash incident.
@Firebase pls help
Dennis Daume retweeted
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
Dennis Daume retweeted
Astra is really, really cool but I cannot current trust it for my present day engineering. At least until I have adjusted. Some random thoughts including code samples of what I recovered from my traces. lucumr.pocoo.org/2026/9/7/as…
Dennis Daume retweeted
That explains why the search in linear is still useless. Maybe ask a real person instead next time 🙈
Dennis Daume retweeted
🎨 new technologies necessitate new interfaces
with real-time latent consistency models, here’s a different kind of drawing app:
Dennis Daume retweeted
Remember back when we all read Clean Code and then to do one thing we wrote 25 small functions across several files instead of one 20-line function that you could look at and say "I know exactly what this function is doing"
Dennis Daume retweeted
Absolutely zero offense to the author, but how can anyone open this test and even begin to understand what's going on here?
It succeeds – what exactly does it verify? It fails – what exactly is broken?
That's the kind of code that mocking frameworks inevitably bring you toward.
Dennis Daume retweeted
10x engineers blah blah blah who cares. The engineers that impress me the most are the ones that fearlessly dive into code, domains they know absolutely nothing about and end up producing amazing work in a short amount of time anyways. Truly awe-inspiring every time I see it.
Dennis Daume retweeted
A common mistake: "Let's abstract this in case we want to replace it later."
Examples:
🚫 "Let's abstract React in case we switch JS frameworks."
🚫 "Let's abstract Tailwind in case we switch CSS approaches."
🚫 "Let's abstract Mongo in case we switch Databases".
This is typically a waste of time, and often hinders current use too.
Why? Because to effectively abstract something, we need to know the API we're abstracting.
But we have no idea what a new technology's API will be. Thus, we have no idea whether our abstraction will work with the new technology.
In summary, abstraction can be useful. But avoid abstracting merely to support changing to some unknown future technology.