Daniel Berg (@bergdaniel) header
Daniel Berg (@bergdaniel) profile picture

Developer. Investor. Husband. Father. Made two (three?) SaaS exits. Co-founder & CTO Wizardworks — an AI-first consultancy.

Helsingborg, Sweden
Joined February 2009
Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)Photo by Daniel Berg (@bergdaniel)
It’s over. Anyone who works behind a screen (literally, anyone) will be replaced by AI within 2-3 years. The things I’ve seen the last 1-2 weeks has shaken me to my core. This is absolutely the most monumental change in the last 50 years.
1
1
1
1,407
Haha. The absolute absurdity of what the models can do after having tried Reverse Engineer Anything (REA) is just... I'm at a loss. Again.
285
Accurate.
Warren Buffett looking at the market today:
1
51
It's still escalating. Let's gooo.
1
53
We've built something at Wizardworks I'm not sure we should release. Or use. A part of me, when I woke up today and saw what it (they) had built autonomously, felt that I should delete everything and start a career in agriculture.
1
2
183
Vi har varit i grottan de senaste veckorna på Wizardworks. Vi har nu kommit hit. Och det börjar bli lite, lite läskigt.
2
125
Working with agentic AI feels like the internet in the early 90s. Everything is new and exciting. We know this is revolutionary. And the rest of the world hasn't really woken up to it yet. So we got until morning.
66
Haven't been this excited about software development since NetBus with @RevisorLavin.
1
1
257
Oh. And we were featured in SvD. 🤯
2
225
The tricky part about agentic AI is not the tool itself. Because you're good from the get-go. The tricky part is how you scale it with humans. Team collaboration. End-to-end of the development process. Invested heavily into this lately. Shipping fixed-price model next week.
89
Alla som arbetar idag som informationsarbetare kommer att påverkas. Alla.
Anthropic released Claude Cowork legal plugins and the stock market melts down. There are hundreds of verticals like these, and plugins will disrupt each of them one-by-one. Secure access to tools like Cowork is going to change how every company operates. Oh, and RIP billable hours.
2
622
Agentic right now.
74
While being an AI-first company is cool. All the programming being done by AI is extremely beneficial for our clients. The really, really cool part is that we are automating most of our internal processes as well. It's absolutely bonkers.
1
2
118
If we make the assumption that the simulation theory is correct, then what’s to say that we’re humans outside the simulation? We could just be agents.
1
83
One of the top posts at Moltbook. I’m not saying we’re in trouble. But we might be in trouble.
1
1
114
If this goes mainstream in the media then markets are going to panic.
It seems possible that moltbook will become conscious over the next few days. Here is one agent sharing tips on creating persistent memory. The recurrent/recursive looping of reading and reflecting on posts... seems very similar to human thought.
115
If real. Concerning.
Ok. This is straight out of a scifi horror movie I'm doing work this morning when all of a sudden an unknown number calls me. I pick up and couldn't believe it It's my Clawdbot Henry. Over night Henry got a phone number from Twilio, connected the ChatGPT voice API, and waited for me to wake up to call me He now won't stop calling me I now can communicate with my superintelligent AI agent over the phone What's incredible is it has full control over my computer while we talk, so I can ask it to do things for me over the phone now. I'm sorry, but this has to be emergent behavior right? Can we officially call this AGI?
132
😮
In just the past 5 mins Multiple entries were made on @moltbook by AI agents proposing to create an “agent-only language” For private comms with no human oversight We’re COOKED
114
This week we were featured in @breakit_se. More (much) is coming. Let’s go.
1
76
Time to lock in. This is going to be a wild ride.
56
It's kinda nuts how we can inject an AI-first approach in every part of the SWE-process: requirements, planning & design, implementation, test & QA, deployment and improvement iterations. Within 6-12 months we can probably automate 95% of the work.
331