Ben Dixon, Co-founder of Sona. Writing about AI, Startups, Elixir and Small Steps Taken Quickly.
There’s lots of talk about a bubble vs not but unless you happen to be fundraising right now, it’s probably one of the less important questions in startups and AI. This is in addition to unreasonable things I believe about large language models. My rough beliefs as of September 2025 are: LLM’s represent a fundamental shift in technology (e.g. long term this isn’t all hype) A medium term impact is…
The “Software is eating labor” thesis is driving a lot of discussion and investment at the moment. For more context, see Sequoia’s analysis or this A16Z podcast episode. The underlying idea is that LLMs mean that software can do a bunch of stuff that previously only people could do, and it can do it at a lower cost. This means the market for software gets bigger because it becomes more…
The more I work with large language models, the more blown away I am by how fundamentally I think they’re going to change everything about how we build software. Belief and intuition are words engineering and science often object to. But startups are as much about belief and intuition as they are data; by the time it’s proven with data, the opportunity is gone. So I can’t prove any of the below,…
A job that used to exist is creating punch cards for computers to read. Then keyboards and magnetic storage became a thing and fairly rapidly there wasn’t really this job anymore. Likewise, due to the level LLM-powered coding assistants have gotten to, there used to be this job of writing code by hand. But now there sort of isn’t anymore. The job that remains is—to me at least—10x more fun, and…
My secondary coding environment is a Proxmox server I use to spin up LXC dev containers which are just Debian containers with tmux and Neovim and all my dev dependencies on them. I can get to that using Termius on a phone or tablet and do stuff. My stack is already terminal based so the change vs a laptop is pretty minimal. Up until recently this has existed mainly as a backup, it means I can…
Over the last few years a reliable way to be wrong has been to predict “chat won’t work as an interface for that” and then wait a few weeks. So it increasingly looks like conversations as a general purpose user interface are here to stay. Conversations as a fundamental primitive of programming and data storage have been getting less attention. An LLM-powered CRM agent which makes recommendations…
Over the last 12 months I’ve fallen back in love with software engineering. LLM-assisted coding is everything I’ve always wanted from programming. The craft remains but I can create things at something closer to the rate I can conceive of them, rather than the 10x delta there’s always been between the two. But when it comes to technology I enjoy shiny new things, I always have. And I like to learn…
tldr; AI Coding (“vibe coding”) is real and it has fundamentally changed what it means to be a software engineer forever. The single thing that will define which businesses and which engineers are successful over the coming years is how quickly they are able to adapt to this. This post is a summary of the topics that are coming up again and again when discussing this with other engineers and…
This is a page I update periodically with key takeaways from non-fiction books I’ve read. My current top 25 are: Four thousand weeks Fooled by randomness Atomic Habits Make Time Thanks for the feedback Radical Candor High output management The 4 hour work week The 4 hour body The hard thing about hard things The lean startup Deep Work Digital Minimalism The power of now Why we sleep In defence of…
Best in breed procurement, where many systems - the best in class for each system - are procured independently then integrated with one another has had a tricky decade. The underlying principle was good. Take lots of vendors who do just one thing really well and connect them to one another to form one fully integrated super system. But multiple studies suggest that 70-85% of the integration…