What Is Sideship and Why Did I Start It?
TLDR: Sideship exists because most AI training for developers is terrible. It is either too theoretical, too generic, or taught by people who do not actually build with AI in production. I started sideship to fix that — we run hands-on workshops led by engineers who use AI in real codebases every day. We teach teams to ship AI themselves, not to depend on consultants forever.
[Anders' take — coming soon]
I started sideship because I kept seeing the same pattern. A company decides they need to "adopt AI." They hire a consulting firm. The consultants come in, build a proof of concept, present some slides, and leave. Three months later, the engineering team is back to square one. The POC is abandoned. The slides are forgotten. Nothing shipped.
The problem was never the technology. The models are good. The tools are mature. The APIs are well-documented. The problem is that developers are expected to adopt a fundamentally new way of working with zero practical training. It is like handing someone a race car and expecting them to compete without driving lessons.
Traditional AI training does not work for developers because:
- It focuses on theory instead of practice. Developers do not need to understand transformer architectures to ship AI features. They need to know how to structure a prompt, handle streaming responses, and test AI outputs
- It is too generic. A frontend developer and a backend developer need completely different AI skills. One-size-fits-all workshops waste everyone's time
- It is taught by people who do not ship code. The best AI training comes from people who build with AI in production every single day, who have hit the edge cases and found the workarounds
Sideship does things differently. Every sideship consultant holds a full-time engineering role outside of consulting. We are not career consultants who read about AI — we are engineers who build with it daily. When we run a workshop, we teach from real experience, not from slides we downloaded last week.
Our workshops are hands-on from minute one. No slides about the history of machine learning. No abstract discussions about ethics. Participants write code, build features, and ship working software during the workshop itself. By the time we leave, the team has not just learned about AI — they have actually used it to build something real.
The model is deliberately designed to make us unnecessary. We do not want long-term consulting engagements. We want to spend a few days with your team, transfer the knowledge and confidence they need, and then get out of the way. If we did our job right, you never need to call us again.
The name "sideship" comes from the idea that AI should be something you ship alongside your existing work — a side-shipped capability that enhances what you already do, not a separate initiative that competes for resources. AI is not a department. It is a tool that every developer should know how to use.
Looking back at the first year, the most rewarding part has been watching teams go from skeptical to confident. Engineers who started our workshops saying "I do not see how AI fits into my work" are now using it every day and wondering how they ever worked without it. That transformation is why sideship exists.