Where AI Is Right Now and Where It's Going
4 min readAn honest read on the current state of AI systems, why Claude Code matters, and what the next two years actually look like.
Here's the honest picture of where things are.
The tools are already good enough. Claude Code can build production systems. MCPs can connect it to your entire stack. Skills can give it a playbook for any repeatable task. The technology to do real, valuable work with AI exists today · not in some future version, not when the next model drops. Right now.
What most people are still missing isn't the tools. It's the thinking.
A lot of the AI content out there right now is focused on building cool automations · things that are technically impressive but don't connect to a real business problem:
- Chatbots that don't get used
- Workflows that save 10 minutes a week
- Demos that impress on LinkedIn but don't deliver ROI
The shift that matters is moving from "what can I build with AI" to "what does this business actually need, and can I quantify the value of fixing it."
The context thesis · why the next two years are going to be about this more than anything else
Right now, there's still a meaningful difference between people who know how to use Claude Code and people who don't. That gap is closing fast. In two years, the tool layer will be fully democratised. Everyone will have access to the same models, the same MCPs, the same skills ecosystem.
The moat won't be tools. It'll be context. Everyone will have the same models and MCPs in two years · the edge goes to whoever fed their tools the better information. That's the whole bet behind your brain.
The businesses and builders who spend the next two years intentionally building structured knowledge · documented processes, second brains, well-organised project context, detailed CLAUDE.md files · will get dramatically better outputs from AI than those who haven't. Not because they have better tools. Because they've given their tools better information to work with.
"2025–2035 is the decade of agents."
Andrej Karpathy, co-founder of OpenAI
Context is infrastructure. The people treating it that way right now are building a compounding advantage that will be very difficult to close later.
SkillTree is built around that idea. Everything here · the skills, the MCPs, the CLAUDE.md philosophy, the audit framework · is context infrastructure.