proud dad of 2 - pwa fanboy - building @syntwinai - organizing Munich TypeScript
München, Bayern
Joined April 2016
- Tweets6.5K
- Following621
- Followers521
- Likes14.9K
Pinned Tweet
i'm looking for our second product engineer in munich (hybrid) at @syntwinai.
syntwins goal is to scale communication capacities through an individual AI counterpart for every person.
instead of leaving a message and wait, you can talk to someones syntwin and often get the answer immediately.
we are an early stage startup with a few investors and B2B customers. the stack is @astrodotbuild, @convex, @livekit and @EffectTS_.
compensation is 90-120k € + equity. find details here:
syntwin.ai/careers/product-e…
if we do eu vs us cars comparisons, can we at least compare roadster vs mission s porsche?
still very representative though.
listen to this man 👇
When I was writing Unreal Engine 1, my time was divided as follows:
- Around 30% went to seriously deep coding on hard problems: assembly language texture mappers, C++ BSP trees, data persistence, object oriented systems design, and networking.
- Around 70% was just typing and refactoring large bodies of code needed for a complete engine feature set. This was not hard, and while I was doing it my mind was mostly on the deeper coding topics, thinking and planning for what's next. It didn't feel much different from vibe coding today, except that it was just me typing code on an IBM Model M keyboard rather than prompting it and reviewing the results.
So, it has always seemed perfectly logical that we would rely on AI for that 70%, once the AI could do it with a similar quality standard as we could, which has never been perfection. It came as no surprise that, after a few years of AI model evolution, it was able to take the reigns on that 70% and free up programmers' time for more valuable tasks.
What's more of a surprise is the emergence of AI as a sort of insta-Carmack able to solve hard technical problems quite cleverly and often cleanly, and that it's able to reason about and work in complex 3D scenes both by operating the tools and by directly understanding the raw data files.
I abandoned assembly language in 1998 when out-of-order CPUs made much simpler C/C++ code faster, and never I missed it. If prompting through architectural documents, concept art, and design plans can transcend typing code and tweaking vertices, that will enable developers to spend their time on higher-level, higher value technical and creative tasks.
What matters is the end product, whether it's a game or an engine. Anything developers can do to try out more ideas and make better software and games faster is a win for everyone.
why wouldnt it just fly there or fly on the roof? its the future of parking if we can get rid of these ramps.
using @ChatGPT for generating images for my recipes makes my recipe list way more beautiful.
here's my image of my favorite porridge vs the one chat made.
we do this at the german military. i had meetings with a peter22mueller and everyone was calling him peter22mueller.
we're hiring our first person with the same name as an existing team member
should we give them [email protected]
"software factory" can be as simple as a skill (or a few) that run on cron jobs or triggered by webhooks. they hand over state in github. done.
what are people building here? i don't get it.
i just had chatgpt shop my groceries and it went really well.
i gave it several old invoices so it knows our usual picks and amounts, e.g. which milk to pick when the list says milk
and access to our recipe + grocery list app
it just did it, did it well, and given 50 items on the list it even did it faster than i would.
it even said "your usual müsli is not in stock, which alternative do you want?"
you cant vibe-code attention for a product.
sure. clone an existing one. or the subset you need. host it yourself and maintain it if you want.
but software that is not taken care of by its creators is doomed. with or without AI.
i rather buy from people who care about their products.
I vibe-coded beehiiv, and it's 83% cheaper.
✨✨ Introducing MailCheap.ai ✨✨
Everything you need to run a newsletter...
✍️ Editor
📅 Scheduling
📈 Stats
🌐 A website for every issue
🛡️ Deliverability guardrails
🔗 Send from your own domain (and keep your reputation)
And it's not a dinosaur tool:
+ It's open source; you can run it for free
+ It's agentic; everything in the UI is available to your AI
Import from beehiiv, Substack, or Mailchimp in 2 minutes.
You pay for the emails you send, not for dormant subscribers. From $19/month.
Oh, and no "Powered By" badge at the bottom of your emails.
after years of listening to rework podcast i finally get to choose the tools myself. and i'm trying @basecamp for our still small team.
we just had slack and github and its a mess. you don't get a picture of whats going on when everything is scattered and mixed with other unrelated stuff.
especially now that there is so much more of everything when agents are involved.
basecamp could be the place where the humans intentionally keep each other up to date with project related work. ideas spark while communicating in slack, and we move to basecamp as soon as someone is working on it.
thats the idea at least. its not straightforward if something should be a message on the message board, or a todolist, or an entire project.
projects feel like too much now that a single person can do really big things in 1-2 days with agents.
how do you guys do it at @37signals?
i brainstorm with the agent
i ask it to pull in another agent to give us outside opinions
then the agent implements something we can use and iterate on
we then discuss where we landed, explore the UX, edge cases, etc
when im about to finalize, i let it explain the implementation to me. and to remove all leftovers from our explorations.
and then it creates a PR with video recordings of the feature and a briefing about how we approached it, making the review really easy.
this is really fun.
