What engineers need to know right now
"I spent two days at AI Native DevCon. Most of the noise around AI is mysticism. Three themes cut through it in a way that's directly relevant to how we build."
What AI actually is, stripped of hype
The new engineering stack it's producing
How your team works inside that stack
"A new kind of software that is good at pattern recognition"
Not magic. Not (yet) thinking. The core idea is 70 years old — what changed was compute + data, not a new algorithm. Every time someone says "the AI thinks X", replace "thinks" with "statistically continues a pattern". It changes how you evaluate risk.
A wolf vs. husky classifier was actually detecting snow in the background. A skin cancer app was detecting doctors' rulers. Silent failures: no error thrown, confident output, wrong answer.
Nobody knows why a trained network works — not even the people who built it. It works until it doesn't. Maintenance is not optional.
"This stack is already here. Claude Code is a harness. CLAUDE.md is context. The question isn't whether we're building on this stack — it's whether we're doing it deliberately."
New compute primitives — the "operating systems"
CLI, MCP, APIs: deterministic arms and legs
Everything that programs the model (rules, skills, passive context)
Deterministic software wrapping the probabilistic model
Composed harnesses: repeatable pipelines
If code programs the machine, context engineering programs the model.
Anthropic's Applied AI team defines it as the primary lever for turning raw model intelligence into "durable, scalable, useful product" — and "a really great investment... [that] has the effect of multiplying the intelligence even as models get smarter."
Context comes in three forms, each solving a different problem:
Skills use progressive disclosure: the agent scans a short frontmatter to decide relevance, then loads the full body only when needed. Think of it as a bookshelf — you scan titles, pull out the book you need.
Context engineering is the discipline of designing all three layers intentionally — what goes in, when it loads, and how it composes. If a skill is wrong, the agent confidently does the wrong thing with perfectly good code underneath.
Up from ~0 at the start of 2025
In one open-source ecosystem alone
"The Humans Architect the System, the AI Writes the Code"
"Vibes don't scale"
Engineers write implementation
Engineers own intent, architecture, and constraints; agents implement
The value moves upstream — into the decisions the agent cannot make.
Behaviour acceptance criteria defined upfront are the source of truth — not a passing CI run, not the agent's confidence.
What the system must do
What must not change; what patterns must be preserved
The test that proves the output is correct
Version them, scan them, eval them, own their lifecycle.
Every agent session starts with a constraint list and UAT criteria — not a vague prompt.
Confident AI output is not correct output. Review gates must check against original intent.
AI Native DevCon: Key Takeaways